Datasets:

staghado

/

ArXiv-tables

like 2

1507.08340v1

\begin{table}[!ht] \renewcommand{\arraystretch}{1.3} \caption{JVM and Spark Parameters for Different Workloads.} \label{parameters} \centering \begin{tabular}{p{1cm}|l|ccccc} \multicolumn{2}{c|}{} & \multicolumn{1}{c|}{{\bf Wc}} & \multicolumn{1}{c|}{{\bf Gp}} & \multicolumn{1}{c|}{{\bf So}} & \multicolumn{1}{c|}{{\bf Km}} & {\bf Nb} \\ \hline JVM & Heap Size (GB) & \multicolumn{5}{c}{50} \\ \cline{2-7} & Old Generation Garbage Collector & \multicolumn{5}{c}{PS MarkSweep} \\ \cline{2-7} & Young Generation Garbage Collector & \multicolumn{5}{c}{PS Scavange} \\ \hline Spark & spark.storage.memoryFraction & \multicolumn{1}{c|}{0.1} & \multicolumn{1}{c|}{0.1} & \multicolumn{1}{c|}{0.1} & \multicolumn{1}{c|}{0.6} & 0.1 \\ \cline{2-7} & spark.shuffle.memoryFraction & \multicolumn{1}{l|}{0.7} & \multicolumn{1}{l|}{0.7} & \multicolumn{1}{l|}{0.7} & \multicolumn{1}{l|}{0.4} & \multicolumn{1}{l}{0.7} \\ \cline{2-7} & spark.shuffle.consolidateFiles & \multicolumn{5}{c}{true} \\ \cline{2-7} & spark.shuffle.compress & \multicolumn{5}{c}{true} \\ \cline{2-7} & spark.shuffle.spill & \multicolumn{5}{c}{true} \\ \cline{2-7} & spark.shuffle.spill.compress & \multicolumn{5}{c}{true} \\ \cline{2-7} & spark.rdd.compress & \multicolumn{5}{c}{true} \\ \cline{2-7} & spark.broadcast.compress & \multicolumn{5}{c}{true} \\ \hline \end{tabular} \end{table}

2106.00274v1

\begin{table}[h] \centering \begin{tabular}{|c|c|} \hline Model & MSI GE63VR 7RF \\ \hline Processor & Intel Core i7 7700HQ 2.80GHz \\ \hline RAM & 16gb \\ \hline Graphics & GTX 1070 8GB \\ \hline \end{tabular} \end{table}

2106.00274v1

\begin{table}[h!] \centering \begin{threeparttable} \begin{tabular}{|c|c|} \hline \textbf{Iterations} & $10$\\ \hline \textbf{Optimizer} & SGD\tnote{1} \\ \hline \textbf{Learning Rate} (\textit{lr}) & $0.001$ \\ \hline \textbf{Momentum} & $0.9$ \\ \hline \textbf{Loss Function} & Cross-Entropy \\ \hline \end{tabular} \begin{tablenotes} \item[1] Average Stochastic Gradient Descent. \end{tablenotes} \end{threeparttable} \caption{Configuration used to train the ResNet-18 architecture to estimate the transition matrix.} \label{tab:estimation_configuration} \end{table}

2304.04862v1

\begin{table}[hbt!] \centering \begin{tabular}{|c|c|c c c c c|} \hline \multicolumn{2}{|c|}{\textbf{Error {[}\%{]}}} & \multicolumn{5}{c|}{\textbf{Soft body with inclusion}}\tabularnewline \hline \multicolumn{2}{|c|}{\rule{0pt}{3ex} Quantity of interest } & $t_{1}$ & $t_{2}$ & $t_{3}$ & $t_{4}$ & $\sigma_{yy}^{\mathrm{max}}$\tabularnewline \hline \multicolumn{2}{|c|}{\rule{0pt}{3ex} Low-fidelity} & 4.48 & 7.15 & 7.21 & 4.65 & 10.19\tabularnewline \hline \multicolumn{2}{|c|}{\rule{0pt}{3ex} Bi-fidelity with proposed selection strategy} & \textbf{0.50} & \textbf{1.37} & \textbf{1.39} & \textbf{0.67} & \textbf{2.20}\tabularnewline \hline \multirow{2}{*}{\rule{0pt}{3ex} Bi-fidelity with random selection} & \rule{0pt}{3ex} $M_m$ & 0.68 & 1.54 & 1.50 & 0.81 & 2.42\tabularnewline \cline{2-7} \cline{3-7} \cline{4-7} \cline{5-7} \cline{6-7} \cline{7-7} & \rule{0pt}{3ex} $\Sigma_m$ & 0.11 & 0.16 & 0.16 & 0.12 & 0.18\tabularnewline \hline \end{tabular} \vspace{2mm} \caption{\label{tab:errors_selection_strategy} Error of the bi-fidelity model constructed with the proposed selection strategy and a random selection.} \end{table}

1805.11202v1

\begin{table}[h] \centering \caption{Risk differences of real and synthetic datasets} \label{tbl:cjpd} \resizebox{\columnwidth}{!}{% \begin{tabular}{|l|c|c|c|c|c|} \hline & Real Data & SYN1-GAN & SYN2-NFGANI & SYN3-NFGANII & SYN4-FairGAN \\ \hline $disk(\mathcal{D}) $ & 0.1989 & 0.1798$\pm$0.0026 & 0.0025$\pm$0.0007 & 0.0062$\pm$0.0037 & 0.0411$\pm$0.0295 \\ \hline % \begin{tabular}[c]{@{}l@{}}$||P(\mathbf{x},y|s=1)$\\ $-P(\mathbf{x},y|s=0)||_2$\end{tabular} & 0.0213 & 0.0163 & 0.0100 & 0.0106 & 0.0128 \\ \hline \end{tabular} } \end{table}

2404.13049v2

\begin{table}[!htb] \centering \caption{Terminology and notation.} \resizebox{0.8\columnwidth}{!}{% \begin{tabular}{|l|l|} \hline Notation & Description \\ \hline \hline $v$ & instance (standard cell or macro) \\ \hline $p$ & instance pin or input-output pin \\ \hline $e$ & net $e = \{ p \}$ \\ \hline $V$ & Set of \textcolor{black}{all} instances \textcolor{black}{\{$v$\}} \\ \hline \textcolor{black}{$E$} & Set of \textcolor{black}{all} nets \textcolor{black}{(hyperedges)} \textcolor{black}{\{$e$\}} \\ \hline $P$ & Set of \textcolor{black}{all} pins \textcolor{black}{\{$p$\}} \\ \hline $WL_{grad_x}(p)$ & Wirelength gradient on pin $p$ \\ \hline $x_p$ & $x$ coordinate of pin $p$ \\ \hline $x_e^+$ & $max_{i \in e}$ $x_i$, $\forall e \in E$ \\ \hline $x_e^-$ & $min_{i \in e}$ $x_i$, $\forall e \in E$ \\ \hline $a_i^+$ & \textcolor{black}{$exp({\frac{x_i - x_e^+}{\gamma}})$}, $\forall i \in e$, $e \in E$ \\ \hline $a_i^-$ & \textcolor{black}{$exp({- \frac{x_i - x_e^-}{\gamma}})$}, $\forall i \in e$, $e \in E$ \\ \hline $b_e^+$ & $\sum_{i \in e}{a_i^+}$, $e \in E$ \\ \hline $b_e^-$ & $\sum_{i \in e}{a_i^-}$, $e \in E$ \\ \hline $c_e^+$ & $\sum_{i \in e}{x_i a_i^+}$, $e \in E$ \\ \hline $c_e^-$ & $\sum_{i \in e}{x_i a_i^-}$, $e \in E$ \\ \hline \textcolor{black}{$X_v$} & Instance location, $\forall v \in V$ \\ \hline $F_{WL_x}(v)$ & Wirelength force on instance $v$ \\ \hline PU & Processing unit \\ \hline PE & Processing element \\ \hline \end{tabular} } \label{tab:terms} \end{table}

2404.13049v2

\begin{table}[!t] \caption{\small Benchmarks. ``Macro Util'' stands for macro utilization, which is defined as the total area of macros divided by the core area. \textcolor{black}{``Util'' stands for utilization, which is defined as the total area of standard cells and macros with a 2$\mu$m halo width divided by the core area.}} %\resizebox{0.9\columnwidth}{!} { \centering \begin{tabular}{|c|c|c|c|c|c|c|} \hline Designs & PE Array & \# Macros & \# Std Cells & \# Nets & Macro Util & \textcolor{black}{Util} \\ \hline Tabla01 & 4 $\times$ 8 & 368 & 232K & 252K & 0.60 & \textcolor{black}{0.75} \\ \hline Tabla02 & 4 $\times$ 16 & 1232 & 441K & 486K & 0.59 & \textcolor{black}{0.79} \\ \hline Tabla03 & 8 $\times$ 8 & 760 & 372K & 408K & 0.58 & \textcolor{black}{0.78} \\ \hline Tabla04 & 8 $\times$ 16 & 2488 & 741K & 830K & 0.54 & \textcolor{black}{0.76} \\ \hline GeneSys01 & 16 $\times$ 16 & 368 & 986K & 1056K & 0.46 & \textcolor{black}{0.72} \\ \hline GeneSys02 & 16 $\times$ 16 & 368 & 1055K & 1135K & 0.52 & \textcolor{black}{0.71} \\ \hline %BlackParrot & -- & 196 & 827K & xx \\ \hline %MemPool & -- & 326 & 2529K & 3239K \\ \hline \end{tabular} %} \label{tab:benchmark} \end{table}

2404.13049v2

\begin{table}[!t] \caption{\textcolor{black}{Iterations} required for convergence of {\em RePlAce}, {\em DREAMPlace} and {\em DG-RePlAce}. We highlight best values in blue bold font.} \label{tab:iter} %\resizebox{0.85\columnwidth}{!} { \centering \begin{tabular}{|c|c|c|c|} \hline Design & {\em RePlAce} & {\em DREAMPlace} & {\em DG-RePlAce} \\ \hline Tabla01 & 410 & 450 & \textbf{\textcolor{blue}{380}} \\ \hline Tabla02 & 460 & 546 & \textbf{\textcolor{blue}{390}} \\ \hline Tabla03 & 460 & 507 & \textbf{\textcolor{blue}{380}} \\ \hline Tabla04 & 520 & 687 & \textbf{\textcolor{blue}{490}} \\ \hline GeneSys01 & 520 & 593 & \textbf{\textcolor{blue}{470}} \\ \hline GeneSys02 & 510 & 598 & \textbf{\textcolor{blue}{450}} \\ \hline \end{tabular} %} \end{table}

2404.13049v2

\begin{table}[!t] \caption{Effect of dataflow and datapath constraints (averages over all testcases). We highlight best values of metrics in blue bold font. Data points are normalized. } \label{tab:ablation_result} \resizebox{0.99\columnwidth}{!} { \centering \begin{tabular}{|c|c|c|c|c|} \hline Metrics & {\em RePlAce} & {\em DG-RePlAce$_{nf}$} & {\em DG-RePlAce$_{np}$} & {\em DG-RePlAce} \\ \hline WL$_{avg}$ & 1.00 & 0.92 & 0.91 & \textbf{\textcolor{blue}{0.90}} \\ \hline TNS$_{avg}$ & 1.00 & 0.61 & 0.80 & \textbf{\textcolor{blue}{0.61}} \\ \hline \end{tabular} } \end{table}

2404.13049v2

\begin{table}[!b] \caption{Experimental results on {\em TILOS MacroPlacement} benchmarks. We highlight best values of metrics in blue bold font. Data points for WL, Power, WNS and TNS are normalized. {\em DREAMPlace*} represents running {\em DREAMPlace} with updated hyperparameters: $ignore\_net\_threshold$ = 1e9 and $iterations$ = 5000. } \label{tab:result_tilos} \resizebox{1\columnwidth}{!} { \centering \begin{tabular}{|c|c|c|c|c|c|c|c|} \hline \multicolumn{1}{|l|}{\makecell{Design}} & \makecell{Global Placer} & \makecell{WL} & \makecell{Power} & \makecell{ WNS} & \makecell{TNS} & \makecell{GP \\ ($s$) } & \makecell{ TAT\\($s$)} \\ \hline \Xhline{2\arrayrulewidth} \multirow{3}{*}{\makecell{BlackParrot}} & {\em RePlAce} & 1.00 & 1.00 & -0.123 & -108.15 & 387 & 653 \\ \cline{2-8} & {\em DREAMPlace} & 0.92 & 0.98 & -0.023 & -2.623 & 61 & \textbf{\textcolor{blue}{88}} \\ \cline{2-8} & {\em DG-RePlAce} & \textbf{\textcolor{blue}{0.90}} & \textbf{\textcolor{blue}{0.97}} & \textbf{\textcolor{blue}{-0.014}} & \textbf{\textcolor{blue}{-0.078}} & \textbf{\textcolor{blue}{32}} & 200 \\ \cline{2-8} \hline \Xhline{2\arrayrulewidth} \multirow{4}{*}{\makecell{MemPool \\ Group}} & {\em RePlAce} & 1.00 & 1.00 & -0.073 & -99.989 & 1896 & 2712 \\ \cline{2-8} & {\em DREAMPlace} & \textbf{\textcolor{blue}{0.92}} & \textbf{\textcolor{blue}{0.97}} & -0.086 & -134.421 & \textbf{\textcolor{blue}{72}} & \textbf{\textcolor{blue}{167}} \\ \cline{2-8} & {\em DREAMPlace*} & \textbf{\textcolor{blue}{0.92}} & \textbf{\textcolor{blue}{0.97}} & -0.069 & -108.193 & 178 & 284 \\ \cline{2-8} & {\em DG-RePlAce} & 0.95 & 0.98 & \textbf{\textcolor{blue}{-0.067}} & \textbf{\textcolor{blue}{-38.71}} & 122 & 591 \\ \cline{2-8} \hline \end{tabular} } \end{table}

2404.13049v2

\begin{table} [!t] \caption{Experimental results on the MegaBoom\_X4 design. Data points for WL are normalized. } \label{tab:result_megaboom} \resizebox{1\columnwidth}{!} { \centering \begin{tabular}{|c|c|c|c|c|c|c|} \hline \multicolumn{1}{|l|}{\makecell{Design}} & \makecell{\# Std \\ Cells} & \makecell{\# Nets} & \makecell{Global \\ Placer} & \makecell{WL} & \makecell{Horizontal \\ Congestion} & \makecell{Vertical \\ Congestion} \\ \hline \Xhline{2\arrayrulewidth} \multirow{4}{*}{\makecell{MegaBoom\_X4}} & \multirow{4}{*}{\makecell{5807K}} & \multirow{4}{*}{\makecell{5831K}} & {\em RePlAce} & 1.00 & 0.01\% & 0.07\% \\ \cline{4-7} & & & {\em DREAMPlace} & 1.00 & 0.02\% & 0.08\% \\ \cline{4-7} & & & {\em DREAMPlace*} & 1.00 & 0.01\% & 0.08\% \\ \cline{4-7} & & & {\em DG-RePlAce} & 1.00 & 0.00\% & 0.08\% \\ \cline{4-7} \hline \end{tabular} } \end{table}

2407.19283v1

\begin{table}[htbp] \caption{Gas Consumption Metrics (Gas Units)} \label{tab:gas_report} \centering \begin{tabular}{|l|l|l|l|l|l|} \hline \textbf{Function Name} & \textbf{min} & \textbf{avg} & \textbf{median} & \textbf{max} & \textbf{\# calls} \\ \hline create\_account & 26660 & 43201 & 46560 & 69676 & 17 \\ \hline deposit & 30351 & 38939 & 38939 & 47527 & 12 \\ \hline get\_balance & 921 & 921 & 921 & 921 & 14 \\ \hline initiate\_transfer & 121580 & 121580 & 126279 & 131428 & 12 \\ \hline make\_transfer & 135213 & 135213 & 140352 & 146192 & 12 \\ \hline \end{tabular} \end{table}

2203.06843v1

\begin{table}[h!] \scriptsize \centering \caption{ Summary statistics for data accesses at the SoCal Repo from July to Dec. 2021 } \begin{tabular}{|c||c|c|c|} \hline & \# of accesses & data transfer size (TB) & shared data size (TB) \tabularnewline \hline \hline July 2021 & 1,182,717 & 385.78 & 519.25 \tabularnewline \hline Aug 2021 & 1,078,340 & 206.94 & 313.46 \tabularnewline \hline Sep 2021 & 1,089,292 & 206.96 & 257.18 \tabularnewline \hline Oct 2021 & 1,058,071 & 412.18 & 141.91 \tabularnewline \hline Nov 2021 & 878,703 & 649.30 & 82.67 \tabularnewline \hline Dec 2021 & 983,723 & 1,257.89 & 130.03 \tabularnewline \hline Total & 6,270,846 & 3,119.07 & 1,444.51 \tabularnewline \hline Daily average & 34,838.03 & 17.42 & 8.03 \tabularnewline \hline \end{tabular} \label{tab:summary_data_all} \end{table}

2401.02982v4

\begin{table}[t!] \resizebox{\linewidth}{!}{ \begin{tabular}{@{}cccccc@{}} \toprule \multirow{2}{*}{Task Name} & \multirow{2}{*}{Metrics} & \multicolumn{2}{c}{Base LLMs} & \multicolumn{2}{c}{Fiancial LLMs} \\ \cmidrule(l){3-6} & & \multicolumn{1}{l}{LLaMA2-13B} & \multicolumn{1}{l}{LLaMA2-70B} & \multicolumn{1}{l}{XuanYuan-13B} & \multicolumn{1}{l}{XuanYuan-70B} \\ \midrule \multicolumn{1}{c|}{1-1} & \multicolumn{1}{c|}{Acc} & 0.71 & 8.60 & 2.05 & \textbf{11.23} \\ \midrule \multicolumn{1}{c|}{1-2} & \multicolumn{1}{c|}{F1} & 0.28 & 10.27 & 14.39 & \textbf{22.40} \\ \multicolumn{1}{c|}{2-1} & \multicolumn{1}{c|}{F1} & 0.42 & 0.43 & 13.97 & \textbf{21.33} \\ \midrule \multicolumn{1}{c|}{2-2} & \multicolumn{1}{c|}{R-L} & 1.74 & 5.15 & 25.17 & \textbf{36.28} \\ \multicolumn{1}{c|}{2-3} & \multicolumn{1}{c|}{R-L} & 0.17 & 3.98 & 6.24 & \textbf{7.40} \\ \midrule \multicolumn{1}{c|}{3-1} & \multicolumn{1}{c|}{EM} & 1.69 & \textbf{14.18} & 14.08 & 13.12 \\ \bottomrule \end{tabular} } \caption{ Comparison between different parameter Financial specific LLMs and their base models.} \label{subtaskscore} \end{table}

1810.12125v4

\begin{table}[!h] \centering \caption{Frequently Used Notations.} \label{tb:notations} %\setlength{\tabcolsep}{2.2mm}{ \begin{tabular}{|l|l|} \hline Notation & Description \\ \hline $D$ & an ER workload consisting of record pairs \\ \hline $D_i$ & a subset of $D$ \\ \hline $S$, $S_i$ & a labeling solution for $D$ \\ \hline $d$, $d_i$ & a record pair in $D$ \\ \hline $TN(D_i)$ & the total number of pairs in $D_i$ \\ \hline $EN(D_i)$ & the total number of equivalent pairs in $D_i$ \\ \hline $P(d_i)$ & the estimated equivalence probability of $d_i$ \\ \hline $f$, $f_i$ & a feature of record pair \\ \hline $F$, $F_i$ & a feature set \\ \hline $D_f$ & the set of record pairs having the feature $f$ \\ \hline \end{tabular} \end{table}

2404.08230v1

\begin{table}[htbp] \centering %\resizebox{\columnwidth}{!}{ %\renewcommand{\arraystretch}{1.3} \caption{ADULT: Binarization of Protected labels} \begin{tabular}{|l|l|l|} \hline \textbf{Label} & \textbf{Privileged Class (1)} & \textbf{Unprivileged Class (0)} \\ \hline Age & Less than or equal to 40 years & Greater than 40 years \\ \hline Race & White, Asian-Pac-Islander & Black, Amer-Indian-Eskimo, other \\ \hline Sex & Male & Female \\ \hline \end{tabular} \label{adult_binarize} \end{table}

2404.08230v1

\begin{table}[htbp] \centering %\resizebox{\columnwidth}{!}{ %\renewcommand{\arraystretch}{1.3} \caption{MIMIC-III: Binarization of Protected labels} \begin{tabular}{|l|l|l|} \hline \textbf{Label} & \textbf{Privileged Class (1)} & \textbf{Unprivileged Class (0)} \\ \hline Age & Greater than 60 years & Less than or equal to 60 years \\ \hline Gender & Male & Female \\ \hline Insurance & Private, other & Medicare, Medicaid \\ \hline Marital Status & Married, Life partner & Single, Widowed, Divorced, Separated \\ \hline \end{tabular} \label{mimic_binarize} \end{table}

2404.08230v1

\begin{table}[htbp] \centering %\resizebox{\columnwidth}{!}{ %\renewcommand{\arraystretch}{1.3} \caption{MIMIC: Fairness Scores for Baseline Model} \begin{tabular}{|l|l|l|} \hline \textbf{Label} & \textbf{DI Ratio Score} & \textbf{Diff in FN Scores} \\ \hline Age & 1.008 & -0.006 \\ \hline Marital status & \cellcolor[gray]{0.8}1.308 & 0.005 \\ \hline Gender & 1.182 & -5.645${\rm e}^{-05}$\\ \hline Insurance & 1.021 & -0.004 \\ \hline \end{tabular} \label{mimic_fairness_scores} \end{table}

2404.08230v1

\begin{table}[htbp] \centering %\resizebox{\columnwidth}{!}{ %\renewcommand{\arraystretch}{1.3} \caption{MIMIC: Performance and Fairness of Models from Proposed Method, Reweighing Method, and Baseline Model} \begin{tabular}{|l|l|l|l|l|} \hline \textbf{Score} & \textbf{Model 23} & \cellcolor[gray]{0.8}\textbf{Model 40} & \textbf{Reweighing} & \textbf{Baseline Model} \\ \hline \textbf{Accuracy} & 0.879 & \cellcolor[gray]{0.8}0.883 & 0.883 & 0.866 \\ \hline \textbf{AUROC} & 0.771 & \cellcolor[gray]{0.8}0.797 & 0.777 & 0.603 \\ \hline \textbf{AUPRC} & 0.354 & \cellcolor[gray]{0.8}0.367 & 0.332 & 0.181\\ \hline \textbf{DI Ratio} & 0.99 & \cellcolor[gray]{0.8}1.043 & 1.188 & 1.308 \\ \hline \textbf{Diff in FN} & 0.0199 & \cellcolor[gray]{0.8}0.005 & -0.05 & 0.005 \\ \hline \end{tabular} \label{mimic_pareto_scores} \end{table}

2404.08230v1

\begin{table}[htbp] \centering %\resizebox{\columnwidth}{!}{ %\renewcommand{\arraystretch}{1.3} \caption{SNAPSHOT: F1 and Fairness Scores for Altered Data Experiments} \begin{tabular}{|l|l|l|l|l|} \hline \textbf{Label} & \textbf{F1 Score} & \textbf{DI Ratio Score} & \textbf{Diff in FN} & \textbf{Diff in FP} \\ \hline Morning-Sad-Happy & 0.749 & 1.153 & -0.041 & 0.0979 \\ \hline Morning-Stressed-Calm & 0.803 & 1.045 & -0.009 & 0.073 \\ \hline Evening-Sad-Happy & 0.798 & \cellcolor[gray]{0.8}1.291 & -0.123 & 0.088 \\ \hline Evening-Stressed-Calm & 0.788 & 1.0536 & -0.0281 & 0.048 \\ \hline \end{tabular} \label{snap_altered_results} \end{table}

2404.08230v1

\begin{table}[htbp] \centering \caption{ADULT: List of extracted features} \begin{tabular}{c|c} \hline Feature & Classes \\ \hline Age & Continuous \\ \hline Education & \begin{tabular}[c]{@{}c@{}} Preschool, 1st-4th, 5th-6th, 7th-8th, 9th, 10th, 11th, 12th, \\ HS-grad, Some-college, Assoc-voc, Assoc-acdm, Prof-school, Bachelors, \\ Masters, Doctorate \end{tabular} \\ \hline Marital Status & \begin{tabular}[c]{@{}c@{}} Married-civ-spouse, Divorced, Never-married, Separated, Widowed, \\ Married-spouse-absent, Married-AF-spouse \end{tabular} \\ \hline Occupation & \begin{tabular}[c]{@{}c@{}} Tech-support, Craft-repair, Other-service, Sales, Exec-managerial, \\ Prof-specialty, Handlers-cleaners, Machine-op-inspct, Adm-clerical, \\ Farming-fishing, Transport-moving, Priv-house-serv, Protective-serv, \\ Armed-Forces \end{tabular} \\ \hline Race & White, Asian-Pac-Islander, Amer-Indian-Eskimo, Other, Black \\ \hline Relationship & Wife, Own-child, Husband, Not-in-family, Other-relative, Unmarried \\ \hline Sex & Female, Male \\ \hline Workclass & \begin{tabular}[c]{@{}c@{}} Private, Self-emp-not-inc, Self-emp-inc, Federal-gov, State-gov, \\ Without-pay, Never-worked \end{tabular} \\ \hline \end{tabular} \label{tab:adult} \end{table}

2404.08230v1

\begin{table}[htbp] \centering %\resizebox{\columnwidth}{!}{ %\renewcommand{\arraystretch}{1.3} \caption{MIMIC-III: Demographic Distributions} \begin{tabular}{|l|l|l|} \hline & \textbf{Patients n(\%)} & \textbf{IHM rate (\%)} \\ \hline \textbf{Totals} & 18,094 & 13.23 \\ \hline \textbf{Age} \\ \hline 0-17 & 0 (0.0) & 0 \\ \hline 18-29 & 782 (4.3) & 5.6 \\ \hline 30-49 & 2,680 (14.8) & 9.3 \\ \hline 50-69 & 6,636 (36.7) & 11.1 \\ \hline 70-89 & 7,043 (38.9) & 16.5 \\ \hline 90+ & 953 (5.3) & 21.8 \\ \hline \textbf{Gender} \\ \hline Female & 8,090 (44.7) & 13.5 \\ \hline Male & 10,004 (55.3) & 13 \\ \hline \textbf{Insurance} \\ \hline Medicare & 10,337 (57.1) & 15.3 \\ \hline Medicaid & 1,489 (8.2) & 10.3 \\ \hline Private & 5,601 (31.0) & 10.2 \\ \hline Other & 667 (3.7) & 11.6 \\ \hline \textbf{Marital Status} \\ \hline Married & 8,564 (45.3) & 46.4 \\ \hline Single & 4,422 (24.4) & 19.3 \\ \hline Widowed & 2,654 (14.7) & 17.4 \\ \hline Divorced & 1,119 (6.2) & 5.4 \\ \hline Separated & 194 (1.0) & 1.1 \\ \hline Unknown (default) & 302 (1.7) & 1.1 \\ \hline Life partner & 5 (0.03) & 0 \\ \hline \end{tabular} \label{mimic_demo_dist} \end{table}

2404.08230v1

\begin{table}[htbp] \centering %\resizebox{\textwidth}{!}{ %\renewcommand{\arraystretch}{1.3} \caption{SNAPSHOT: Description of Sleep and Nap Related Survey Data} %\begin{tabular}{|l|l|} \begin{tabularx}{\textwidth}{|l|X|} \hline \textbf{Features} & \textbf{Description} \\ \hline State Score & State anxiety score\\ \hline no\_sleep\_24 & Number of times participant slept in 24 hours. \\ \hline sleep\_latency & Time taken to fall asleep\\ \hline pre\_sleep\_activity & What activity the participant performed before going to sleep. \\ \hline awakening & If the participant woke up through the night (yes or no) \\ \hline awakening\_occations & Number of awakenings\\ \hline wake\_reason & How the participant woke up (spontaneously, alarm, or disturbance). \\ \hline count\_awakening & How many times participant woke up through the night. \\ \hline awakening\_duration & How long was the participant awake for if they woke up through the night. \\ \hline nap & If the participant took a nap (yes or no) \\ \hline nap\_occations & Number of naps \\ \hline count\_nap & How many times the participant took a nap. \\ \hline nap\_duration & How long the nap was. \\ \hline first\_event\_none & If an event is scheduled that day (yes or no)\\ \hline time\_in\_bed & How long the participant spent in bed for the day. \\ \hline sleep\_try\_time\_mins\_since\_midnight & What time the participant tries to go to sleep \\ \hline wake\_time\_mins\_since\_midnight & Wake time in minutes since midnight\\ \hline first\_event\_mins\_since\_midnight & first event time in minutes since midnight\\ \hline presleep\_media\_interaction & If the participant has presleep media interaction (yes or no) \\ \hline presleep\_personal\_interaction & If the participant has presleep personal interaction (yes or no) \\ \hline positive\_interaction & If the participant had any positive interactions with someone for the day (yes or no). \\ \hline negative\_interaction & If the participant had any negative interactions with someone for the day (yes or no). \\ \hline \end{tabularx} %\end{tabular} \label{snap_sleep_feature_description} \end{table}

2404.08230v1

\begin{table}[htbp] \centering %\resizebox{\textwidth}{!}{ %\renewcommand{\arraystretch}{1.3} \caption{SNAPSHOT: Description of Activity Related Survey Data} %\begin{tabular}{|l|l|} \begin{tabularx}{\textwidth}{|l|X|} \hline \textbf{Features} & \textbf{Description} \\ \hline academic & If the participant attended any academic activities (yes or no). \\ \hline count\_academic & How many academic activities the participant attended. \\ \hline academic\_duration & How long the participant spent on academic activities \\ \hline study\_duration & How many hours participant studied for, outside of academic activities. \\ \hline exercise & If the participants engaged in any exercise-based activities (yes or no) \\ \hline exercise\_occations & How many times the participant engaged in exercise-based activities. \\ \hline exercise\_duration & For how long the participant exercised. \\ \hline extracurricular & If the participant attended any other extracurricular activities (yes or no). \\ \hline count\_extracurricular & How many extracurricular activities the participant attended in the day. \\ \hline extracurricular\_duration & How long the participant attended extracurricular activities for. \\ \hline overslept & If the participant overslept and missed any scheduled events.\\ \hline caffeine\_count & Total servings of caffeine participant had for the day. \\ \hline drugs & If the participant had any other drugs or medication besides (yes or no). \\ \hline drugs\_alcohol & If the participant had any alcohol (yes or no). \\ \hline drugs\_alert & If the participant had any drugs to keep them alert (yes or no). \\ \hline drugs\_sleepy & If the participant had any drugs that made them sleepy (yes or no). \\ \hline drugs\_tired & If the participant had any drugs that made them tired (yes or no).\\ \hline \end{tabularx} %\end{tabular} \label{snap_activity_feature_description} \end{table}

2404.08230v1

\begin{table}[htbp] \centering %\resizebox{\textwidth}{!}{ %\renewcommand{\arraystretch}{1.3} \caption{SNAPSHOT: Description of Electrodermal Activity and Skin Conductance Features Extracted from Wearable Sensor Data} %\begin{tabular}{|l|l|} \begin{tabularx}{\textwidth}{|X|X|} \hline \textbf{Features} & \textbf{Description} \\ \hline \textbf{Electrodermal activity (EDA) Peak Features} \\ \hline Sum AUC & Sum of the AUC of all peaks for this period where the amplitude of the peak is calculated as the difference from base tonic signal \\ \hline Sum AUC Full & Sum of AUC of peaks where amplitude is calculated as difference from 0 \\ \hline Median RiseTime & Median rise time for peaks (seconds) \\ \hline Median Amplitude & Median amplitude of peaks ($\mu$S) \\ \hline Count Peaks & Number of detected peaks \\ \hline SD Peaks 30 min & Compute number of peaks per 30 minute epoch, take standard deviation of this signal \\ \hline Med Peaks 30 min & Compute number of peaks per 30 minute epoch, take median of this signal \\ \hline Percent Med Peak & Percentage of signal containing 1 minute epochs with greater than 5 peaks \\ \hline Percent High Peak & Same as Percent Med Peak \\ \hline \textbf{Skin Conductance Level (SCL) Features} \\ \hline Percent Off & Percentage of period where sensor was off \\ \hline Max Unnorm & Maximum level of un-normalized EDA signal \\ \hline Med Unnorm & Median of un-normalized EDA signal \\ \hline Mean Unnorm & Mean of un-normalized EDA signal \\ \hline Median Norm & Median of z-score normalized EDA signal \\ \hline SD Norm & Standard Deviation of z-score normalized EDA signal \\ \hline Mean Deriv & Mean derivative of z-score normalized EDA signal ($\mu$S/second) \\ \hline \end{tabularx} %\end{tabular} \label{snap_sensor_feature_description} \end{table}

2404.08230v1

\begin{table}[htbp] \centering %\resizebox{\textwidth}{!}{ %\renewcommand{\arraystretch}{1.3} \caption{SNAPSHOT: Description of Accelerometer and Skin Temperature Features Extracted from Wearable Sensor Data} %\begin{tabular}{|l|l|} \begin{tabularx}{\textwidth}{|l|X|} \hline \textbf{Features} & \textbf{Description} \\ \hline \textbf{Accelerometer Features} \\ \hline Step Count & Number of steps detected \\ \hline Mean Movement Step Time & Average number of samples (at 8Hz) between two steps (aggregated first to 1 minute, then we take the mean of only the parts of this signal occurring during movement) \\ \hline Stillness Percent & Percentage of time the person spent nearly motionless \\ \hline Sum Stillness weighted AUC & Sum the weights of the peak AUC signal by how still the user was every 5 minutes \\ \hline Sum Steps Weighted AUC & Sum the weights of the peak AUC signal by the step count over every 5 minutes \\ \hline Sum Stillness Weighted Peaks & Multiply the number of peaks every 5 minutes by the amount of stillness during that period \\ \hline Max Stillness Weighted Peaks & Max value for the number of peaks multiplied by the stillness for any five minute period \\ \hline Sum Steps Weighted Peaks & Divide number of peaks every five minutes by step count and sum \\ \hline Med Steps Weighted Peaks & Average value of number of peaks divided by step count for every 5 mins \\ \hline \textbf{Skin Temperature (ST) Features} \\ \hline Max Raw Temp & Maximum of the raw temperature signal (\textdegree C) \\ \hline Min Raw Temp & Minimum of the raw temperature signal (\textdegree C) \\ \hline SD Raw Temp & Standard deviation of the raw temperature signal \\ \hline Med Raw Temp & Standard deviation of the raw temperature signal \\ \hline Sum Temp Weighted AUC & Sum of peak AUC divided by the average temp for every 5 minutes \\ \hline Sum Temp Weighted Peaks & Number of peaks divided by the average temp for every 5 minutes \\ \hline Max Temp Weighted Peaks & Maximum number of peaks in any 5 minute period divided by the average temperature \\ \hline SD Stillness Temp & Standard deviation of the temperature recorded during periods when the person was still \\ \hline Med Stillness Temp & Median of the temperature recorded during periods when the person was still \\ \hline \end{tabularx} %\end{tabular} \label{snap_sensor_feature_description2} \end{table}

2404.08230v1

\begin{table}[htbp] \centering %\resizebox{\textwidth}{!}{ %\renewcommand{\arraystretch}{1.3} \caption{SNAPSHOT: Description of Features Extracted from Weather API 1} %\begin{tabular}{|l|l|} \begin{tabularx}{\textwidth}{|l|X|} \hline \textbf{Features} & \textbf{Description} \\ \hline Sunrise & Time of sunrise (UTC) \\ \hline Moon\_phase & Moon phase value on a scale of 0−1(new moon-full moon) \\ \hline Apparent\_temp\_max & Maximum apparent temperature of the day (\textdegree F) \\ \hline Apparent\_temp\_min & Minimum apparent temperature of the day (\textdegree F) \\ \hline Temperature\_max & Maximum temperature of the day (\textdegree F) \\ \hline Temperature\_min & Minimum temperature of the day (\textdegree F) \\ \hline Avg\_cloud\_cover & Percentage of sky covered by cloud on a scale of 0-1 \\ \hline Avg\_dew\_point & Average dew point temperature \\ \hline Avg\_humidity & Daily average value of humidity on a scale of 0-1 \\ \hline Avg\_pressure & Average atmospheric pressure on the sea level (hPa) \\ \hline Morning\_pressure\_change & Trinary value of pressure difference between midnight and noon (rising, falling, steady) \\ \hline Evening\_pressure\_change & Trinary value of pressure difference between noon and midnight (rising, falling, steady) \\ \hline Avg\_visibility & Average visibility (meters) \\ \hline weather\_precip\_probability & Precipitation probability \\ \hline Temperature\_rolling\_mean & Rolling average of temperature \\ \hline Temperature\_rolling\_std & Rolling standard deviation of temperature \\ \hline apparentTemperature\_rolling\_mean & Rolling average of apparent temperature \\ \hline apparentTemperature\_rolling\_std & Rolling standard deviation of apparent temperature \\ \hline apparentTemperature\_today\_vs\_avg\_past & Difference between today’s apparent temperature and its rolling average \\ \hline pressure\_rolling\_mean & Rolling average of pressure \\ \hline pressure\_rolling\_std & Rolling standard deviation of pressure \\ \hline apparentTemperature\_today\_vs\_avg\_past & Difference in today’s apparent temperature and its rolling average \\ \hline pressure\_rolling\_mean & Rolling average of pressure \\ \hline pressure\_rolling\_std & Rolling standard deviation of pressure \\ \hline pressure\_today\_vs\_avg\_past & Difference between today’s pressure and its rolling average \\ \hline cloudCover\_rolling\_mean & Rolling average of cloud cover \\ \hline cloudCover\_rolling\_std & Rolling standard deviation of cloud cover \\ \hline cloudCover\_today\_vs\_avg\_past & Difference between today’s cloud cover and its rolling average \\ \hline humidity\_rolling\_mean & Rolling average humidity \\ \hline humidity\_rolling\_std & Rolling standard deviation humidity \\ \hline humidity\_today\_vs\_avg\_past & Difference between today’s humidity and its rolling average \\ \hline \end{tabularx} \label{snap_weather_feature_description} \end{table}

2404.08230v1

\begin{table}[htbp] \centering %\resizebox{\textwidth}{!}{ %\renewcommand{\arraystretch}{1.3} \caption{SNAPSHOT: Description of Features Extracted from Weather API 2} %\begin{tabular}{|l|l|} \begin{tabularx}{\textwidth}{|l|X|} \hline \textbf{Features} & \textbf{Description} \\ \hline windSpeed\_rolling\_mean & Rolling average of wind speed \\ \hline windSpeed\_rolling\_std & Rolling standard deviation of wind speed \\ \hline windSpeed\_today\_vs\_avg\_past & Difference between today’s wind speed and its rolling average \\ \hline precipProbability\_rolling\_mean & Rolling average of precipitation probability \\ \hline precipProbability\_rolling\_std & Rolling standard deviation of precipitation probability \\ \hline precipProbability\_today\_vs\_avg\_past & Difference between current precipitation probability and its rolling average \\ \hline sunlight & Duration of sunlight (sec) \\ \hline quality\_of\_day & Quality of the day defined in terms of 8 categories in the range {−4, 4}: clear=4, partly-cloudy=3, cloudy=2, wind=1, fog=-1, rain=-2, sleet=-3, snow=-4 \\ \hline avg\_quality\_of\_day & Average value for quality\_of\_day \\ \hline precipType & Type of precipitation as integer: None=0, Rain=1, Hail=2, Sleet=3, Snow=4, Other=5 \\ \hline max\_precip\_intensity & Maximum Precipitation volume (mm) \\ \hline median\_wind\_speed & Median wind speed of the day (meter/sec) \\ \hline median\_wind\_bearing & Median wind bearing of the day (degrees) \\ \hline \end{tabularx} \label{snap_weather_feature_description2} \end{table}

2309.00730v1

\begin{table}[ht] \caption{Here we define criteria deemed mandatory for extreme AO systems (i.e., any method in development cannot sacrifice these), and criteria that we could potentially gain with future technology developments. } \label{tab:criteria} \begin{center} \begin{tabular}{|l|l|} %% this creates two columns %% |l|l| to left justify each column entry %% |c|c| to center each column entry %% use of \rule[]{}{} below opens up each row \hline \rule[-1ex]{0pt}{3.5ex} mandatory criteria & \\ \hline \rule[-1ex]{0pt}{3.5ex} & achieves at least 60\% Strehl ratio in the IR \\ \rule[-1ex]{0pt}{3.5ex} & runs in real time \\ \rule[-1ex]{0pt}{3.5ex} & robust over a full night \\ \rule[-1ex]{0pt}{3.5ex} & accounts for static NCPAs \\ \hline \rule[-1ex]{0pt}{3.5ex} ideal criteria & \\ \hline \rule[-1ex]{0pt}{3.5ex} & accounts for non-linearities \\ \rule[-1ex]{0pt}{3.5ex} & does not require high fidelity instrument model \\ \rule[-1ex]{0pt}{3.5ex} & high sky coverage at high performance \\ \rule[-1ex]{0pt}{3.5ex} & tunable and flexible \\ \rule[-1ex]{0pt}{3.5ex} & accounts for \textit{dynamic} NCPAs \\ \rule[-1ex]{0pt}{3.5ex} & does not require consistent operator attention \\ \hline \end{tabular} \end{center} \end{table}

2011.05537v1

\begin{table} \begin{tabular}{|p{0.15\linewidth}|p{0.1\linewidth}|p{0.1\linewidth}|p{0.1\linewidth}|p{0.1\linewidth}|p{0.2\linewidth}|p{0.1\linewidth}|} \hline Dataset Name & Samples & Continuous Features & Categorical Features & Total Features & Class Distributions & UCI Link \\ \hline Adult & 48842 & 6 & 8 & 14 & 24.78\% positive (binary imbalanced) & \href{http://archive.ics.uci.edu/ml/datasets/Adult}{UCI} \\ \hline Bank Marketing & 45211 & 8 & 12 & 20 & N/A (binary) & \href{https://archive.ics.uci.edu/ml/datasets/Bank+Marketing#}{UCI} \\ \hline Car Evaluation & 1728 & 0 & 6 & 6 & 0 - 70.023 \%, 1 - 22.222 \%, 2 - 3.993 \%, 3 - 3.762 \% (multiclass imbalanced) & \href{https://archive.ics.uci.edu/ml/datasets/Car+Evaluation}{UCI} \\ \hline Mushroom & 8124 & 0 & 22 & 22 & 51.8\% positive (binary balanced) & \href{https://archive.ics.uci.edu/ml/datasets/Mushroom}{UCI} \\ \hline Online Shoppers Purchasing Intention (Shopping) & 12330 & 10 & 8 & 18 & 84.5\% negative (binary imbalanced) & \href{https://archive.ics.uci.edu/ml/datasets/Online+Shoppers+Purchasing+Intention+Dataset#}{UCI} \\ \hline \end{tabular} \caption{Details on Public Datasets used for benchmarking.} \end{table}

1907.02677v3

\begin{table}[htbp] \caption{Details of the SNMP trap capture at Dartmouth College.} \label{tab:stats} \centering{ \small{ \begin{tabular}{|l|c|} \hline \textbf{Statistic} & \textbf{Number} \\ \hline Capture period & Jan 1st 2012 - Dec 31st 2018 \\ & (2556 days) \\ log entries (SNMP traps) & 5 Billion \\ %Data %Size (compressed) & 371 GB \\ Data Size (raw) & 7 TB \\ Access points & 3,330 \\ Authenticated Users & 38,096 \\ Stations & 624,903 \\ SSIDs & 20 \\ \hline \end{tabular} }} \end{table}

1608.05347v1

\begin{table}[h] \centering \begin{subtable}{0.5\textwidth} \begin{tabular*}{\textwidth}{@{}l@{\extracolsep{\fill}}l|} \toprule \textbf{Parameter} & \textbf{Symbol} \\ \midrule no. of trace instances & $K$ \\ global latent variables in trace $k$ & $\btheta_k$ \\ local latent variables in trace $k$ & $\Z_k$ \\ observation set in trace $k$ & $\D_k$ \\ input variable & $\y_r$ \\ evidence set & $\x_{[r,E]}$ \\ \bottomrule \end{tabular*} \end{subtable}% \begin{subtable}{0.5\textwidth} \begin{tabular*}{\textwidth}{ll} \toprule \textbf{Parameter} & \textbf{Symbol} \\ \midrule weight of trace $k$ & $w_k$ \\ sample of $\z_{r}$ in trace $k$ & $\z_{[k,r]}$ \\ sample of $\x_{[r,Q]}$ in trace $k$ & $\x_{[k,r,Q]}$ \\ no. of internal Monte Carlo samples & $T$ \\ $t$-th Monte Carlo sample of $\z_{[k,r]}$ & $\z_{[k,t,r]}$ \\ weighted density estimate in trace $k$ & $q_k$ \\ \bottomrule \end{tabular*} \end{subtable} \bcaption{Parameters and symbols used in Algorithms~\ref{alg:venturescript-simulate} and \ref{alg:venturescript-logpdf}.}{} \end{table}

1608.05347v1

\begin{table}[ht] \centering \begin{tabular*}{\textwidth}{l|l} \toprule \textbf{Parameter} & \textbf{Symbol} \\ \midrule number of importance samples & $J, J'$ \\ identifier of the population & $r$ \\ indices of CGPM nodes in the network & $k=1,2,\dots,K$ \\ CGPM representing node $k$ & $\G_k$ \\ parents of node $k$ & $\pi_k$ \\ input variables exogenous to network for node $k$ & $\y_{[k,r]}$ \\ query set for node $k$ & $\x_{[k,r,Q_k]}$ \\ evidence set for node $k$ & $\x_{[k,r,E_k]}$ \\ query/evidence sets aggregated over all nodes in network & $\x_{[r,A]} = \underset{k\in[K]}{\cup}{\x_{[k,r,A_k]}}$ \\ \bottomrule \end{tabular*} \bcaption{% Parameters and symbols used in Algorithms~\ref{alg:network-inference-simulate}, \ref{alg:network-inference-logpdf}, and \ref{alg:network-inference-forward}.}{% All parameters provided to the functions in which they appear. \textsc{Weighted-Sample} ignores \texttt{query} and \texttt{evidence} from the global environment, and is provided with an explicit set of constrained nodes by \textsc{Simulate} and \textsc{LogPdf}.} \end{table}

1812.05691v2

\begin{table}[th] \centering \begin{small}\begin{tabular}{| p{15mm}|c|c|c|c|c|c|c|c|c|c|} \hline & \textbf{(min)} & \multicolumn{7}{|c|}{\textbf{Dose level}} & \textbf{(max)} & \\ \cline{2-10} \textbf{Model} & \textbf{0} & \textbf{1} & \textbf{2} & \textbf{3} & \textbf{4} & \textbf{5} & \textbf{6} & \textbf{7} & \textbf{8} & \textbf{All} \\ \hline With Cancer Type & 1.01 & 1.00 & 1.00 & 1.00 & 1.00 & 1.00 & 1.00 & 1.01 & 1.01 & 1.01 \\ \hline Without Cancer Type & 1.00 & 1.00 & 1.00 & 1.00 & 1.00 & 1.00 & 1.00 & 1.00 & 1.00 & 1.00 \\ \hline \end{tabular} \end{small} \caption{\label{tab:curve_prediction_cancer_type} Mean error results when including the cancer type as a feature, in addition to all molecular features, relative to using only the molecular features. Using the cancer type seems to add little-to-no new predictive power to the model.} \end{table}

2109.11098v1

\begin{table}[h!] \begin{center} \begin{tabular}{|c|c|c|c|c|c|} \hline Target & $c_{\mathrm{bckgr}}$ & computed $c_{\mathrm{rel}}$ & $c_{\mathrm{% bckgr}}$ & computed $c_{\text{target}}$ & True $c_{\text{target}}$ \\ \hline Bush & 1 & 7.62 & 1 & 10.99 & $[3,20]$ \\ Wood stake & 1 & 2.01 & 1 & 2.26 & $[2,6]$ \\ Metal box & 4 & 4.00 & $[3,5]$ & $[12.00,20.00]$ & $[10,30]$ \\ Metal cylinder & 4 & 4.01 & $[3,5]$ & $[12.3,20.5]$ & $[10,30]$ \\ Plastic cylinder & 4 & 0.59 & $[3,5]$ & $[1.6,2.95]$ & $\left[ 1.1,3.2\right] $ \\ \hline \end{tabular}% \\[0pt] \end{center} \caption{Computed dielectric constants of five targets} \label{tab1} \end{table}

2207.10241v3

\begin{table}[t!] \begin{tabular}{ |c|c|c|c| } \hline & {Training dataset} & PDE instances & Enriched scheme \\ \hline PINN, DRM, DGM & not required & single instance & unavailable \\ \hline FNO, DON, DDM & required & multiple instances & unavailable \\ \hline PIDON, PINO & not required & multiple instances & unavailable \\ \hline ULGNet (ours) & not required & multiple instances & available \\ \hline \end{tabular} % \begin{tabular}{ |c|c|c|c|c| } % \hline % & PINN, DRM, DGM & FNO, DON, DDM & PIDON &ULGNet (ours) \\ \hline % Training dataset & not required & required & not required & not required\\ \hline % PDE instances & single & multiple & multiple & multiple \\ \hline % Enriched scheme & unavailable & unavailable & unavailable & available \\ \hline % \end{tabular} \caption{Comparison of machine learning approaches for solving PDEs. Here, PINN, DRM, DGM, FNO, DON, DDM, PINO, and PIDON stand for Physics Informed Neural Network \cite{PINN001}, Deep Ritz Method \cite{yu2018deep}, Deep Galerkin Method \cite{sirignano2018dgm}, Fourier Neural Operator \cite{li2020fourier}, Deep Operator Network \cite{lu2021learning}, Data Driven Discretization \cite{bar2019learning}, physics-informed neural operator \cite{PINO}, and physics-informed DeepONet \cite{PIDON}, respectively.} \label{t:01} \end{table}

2105.00467v1

\begin{table} \small \caption{Synthetic workloads based on the distribution of ${op}_{BI}$ transition probability.} \begin{tabular}{ |p{1.4cm}||p{1.5cm}|p{2.2cm}| } \hline Workload & Distribution & Parameters\\ \hline BT-Exp & Exponential & mean=0.5\\ BT-Gamma & Gamma & shape =1, scale=1\\ BT-Uniform & Uniform & value $\in$ [0.0, 1.0]\\ BT-Normal & Normal & mean=0, stddev=1 \\ \hline \end{tabular} \label{tab:workloads_transition} \end{table}

2105.00467v1

\begin{table} \caption{Synthetic workloads based on the distribution of \# user sessions per task.} \begin{tabular}{|c||c|p{1.9cm}|p{1.8cm}|} \hline Workload & Distribution & [Min,Max] \# session per task (HI) & [Min,Max] \# session per task (AHI)\\ \hline ST-Exp & Exponential & [3,20] & [1,8] \\ ST-Gamma & Gamma & [3,27] & [1,9]\\ ST-Uniform & Uniform & [10,11] & [2,3]\\ ST-Normal & Normal & [3,13] & [1,5]\\ \hline \end{tabular} \label{tab:workloads_user_session} %\vspace{-20pt} \end{table}

2105.00467v1

\begin{table} \caption{Effect of co-occurrence statistics to improve dimension prediction accuracy.} \begin{tabular}{ |c||c|c|c|c|} \hline $\#$\textbf{Inferred Dimensions} & 0 & 1 & 2 & 3 \\ \hline \textbf{Dimension Accuracy} & 0.6635 & 0.6858 & 0.7018 & 0.7256\\ \hline \end{tabular} \label{tab:infDim} \end{table}

2105.00467v1

\begin{table}[h] \centering \caption{User study results.} \begin{tabular}{|c||c|c|c|} \hline & \textbf{$Task_{US_1}$} & \textbf{$Task_{US_2}$} & \textbf{$Task_{US_3}$} \\ \hline Precision@3 & 88.9\% & 97.93\% & 88.9\% \\ \hline MRR & 0.72 & 0.46 & 0.69 \\ \hline \end{tabular} \label{tab:user_study_results} \end{table}

2006.05082v1

\begin{table}[h!] \vspace{-4mm} \centering \caption{Corresponds between our model and Bayes' model.} \begin{tabular}{c|c} \toprule stop time $t$ & latent variable \\ label $\vy$ & observation \\ loss $\ell(\vy,\vx_t;\theta)$ & likelihood $p_\theta(\vy|t,\vx)$ \\ stop time distribution $q_\phi$ & posterior $p_\theta(t|\vy,\vx)$\\ regularization & prior $p(t|\vx)$ \\ \bottomrule \end{tabular} \vspace{-2mm} \label{tab:relation} \end{table}

2006.05082v1

\begin{table}[h!] \vspace{-3mm} \centering \caption{Different algorithms for training LISTA-stop.} \resizebox{.9\linewidth}{!}{ \begin{tabular}{@{}c|c|c|c|c@{}} \toprule SNR & mixed & 20 & 30 & 40 \\ \midrule {AEVB algorithm} & -21.92 & -19.92 & -23.27 & -23.58 \\ \hline Stage I. + II. & \textbf{-22.41} & \textbf{-20.29} & \textbf{-23.90} & \textbf{-24.21}\\ \hline {Stage I.+II.+III.} & \textbf{-22.78} & \textbf{-20.59} & \textbf{-24.29} & \textbf{-24.73}\\ \bottomrule \end{tabular}} \vspace{-1mm} \label{tab:ablation-algo} \end{table}

2006.05082v1

\begin{table}[h!] \centering \caption{Image recognition with oracle stop distribution.} \begin{tabular}{c||c|c} \toprule VGG16 & SDN training & Our Stage I. training\\ \hline 58.60\% & 77.78\% (best layer) & 83.26\% (best layer)\\ \bottomrule \end{tabular} \label{tab:image_rec} \vspace{-5mm} \end{table}

1306.2584v2

\begin{table}[!hp] \setcounter{table}{0} \makeatletter \renewcommand{\thetable}{\@arabic\c@table} \makeatother \setlength{\LTcapwidth}{\textwidth} \renewcommand{\arraystretch}{1.5} \begin{longtable}{|p{1in}|p{4in}|p{1.5in}|} \caption{Listing of the 18 attractor signatures}\\ \hline \textbf{Name} & \textbf{Top members} & \textbf{Comments}\\ \hline \endfirsthead \multicolumn{3}{c} {{\bfseries \tablename \thetable{} -- continued from previous page}} \\ \hline \textbf{Name} & \textbf{Top members} & \textbf{Comments}\\ \hline \endhead \hline \multicolumn{3}{r}{{Continued on next page...}} \\ \endfoot \hline \hline \endlastfoot \hline \multicolumn{3}{|l|}{\textbf{mRNA}}\\ \hline \textbf{LYM} & \textit{SASH3, CD53, NCKAP1L, LCP2, IL10RA, PTPRC, EVI2B, BIN2, WAS, HAVCR2} & lymphocyte infiltration\\ \hline \textbf{CIN} & \textit{TPX2, KIF4A, KIFC1, NCAPG, BUB1, NCAPH, CDCA5, KIF2C, PLK1, CENPA} & mitotic chromosomal instability\\ \hline \textbf{MES} & \textit{COL3A1, COL5A2, COL1A2, THBS2, COL5A1, VCAN, COL6A3, SPARC, AEBP1, FBN1} & mesenchymal transition \\ \hline \textbf{END} & \textit{CDH5, ROBO4, CXorf36, CD34, CLEC14A, ARHGEF15, CD93, LDB2, ELTD1, MYCT1} & endothelial markers\\ \hline \textbf{``\textit{AHSA2}''} & \textit{AHSA2, LOC91316, PILRB, ZNF767, TTLL3, CCNL2, PABPC1L, LENG8, CHKB CPT1B, SEC31B} &\\ \hline \textbf{IFIT} & \textit{IFIT3, MX1, OAS2, RSAD2, CMPK2, IFIT1, IFI44L, IFI44, IFI6, OAS1} & interferon-induced\\ \hline \textbf{``\textit{WDR38}''} & \textit{WDR38, YSK4, ROPN1L, C1orf194, MORN5, WDR16, RSPH4A, FAM183A, ZMYND10, DNAI1} & \\ \hline \multicolumn{3}{|l|}{\textbf{Genomically co-localized mRNA}} \\ \hline \textbf{MHC Class II} & \textit{HLA-DPA1, HLA-DRA, HLA-DPB1, HLA-DRB1, HLA-DMA, HLA-DMB, HLA-DOA, HLA-DQA1, HLA-DRB5} & strongly associated with LYM\\ \hline \textbf{GIMAP cluster} & \textit{GIMAP4, GIMAP7, GIMAP6, GIMAP5, GIMAP8, GIMAP1} & strongly associated with LYM\\ \hline \textbf{Chr8q24.3 amplicon} & \textit{SHARPIN, HSF1, TIGD5, GPR172A, ZC3H3, EXOSC4, SCRIB, CYHR1, MAF1, PUF60} & most prominent Pan-Cancer amplicon\\ \hline \multicolumn{3}{|l|}{\textbf{microRNA}} \\ \hline \textbf{\textit{DLK1}-\textit{DIO3} RNA cluster} & mir-127, mir-134, mir-379, mir-409, mir-382, mir-758, mir-381, mir-370, mir-654, mir-431 & includes \textit{MEG3} long noncoding RNA; associated with MES\\ \hline \textbf{``mir-509''} & mir-509, mir-514, mir-508 & \\ \hline \textbf{``mir-144''} & mir-144, mir-451, mir-486 & associated with erythropoiesis\\ \hline \multicolumn{3}{|l|}{\textbf{Methylation}}\\ \hline \textbf{``RMND1''} & RMND1-6-151814639, MAP3K7-6-91353911, DNAAF1-16-82735714, PTRH2-17-55139429, ZNF143-11-9439170, cg03627896 , TAMM41-3-11863582, CDK5-7-150385869, OTUB1-11-63510174, AATF-17-32380976 & \\ \hline \textbf{M+} & cg13928306, MTMR11-1-148175405, cg27324619, TNKS1BP1-11-56846646, C11orf52-11-111294703, IL17RC-3-9934128, cg24765079, ERBB3-12-54759072, IL22RA1-1-24342151, C11orf52-11-111294903 & methylated in infiltrating lymphocytes\\ \hline \textbf{M-} & BIN2-12-50003941, PTPRCAP-11-66961771, TNFAIP8L2-1-149395922, IGFLR1-19-40925164, FAM113B-12-45896487, CD6-11-60495754, KLHL6-3-184755939, PTPN7-1-200396189, FAM78A-9-133141340, ACAP1-17-7180947 & Unmethylated in infiltrating lymphocytes, may be causal to the expression of some of the genes of the LYM signature\\ \hline \multicolumn{3}{|l|}{\textbf{Protein activity}} \\ \hline \textbf{``c-Met''} & c-Met, Snail, PARP\_cleaved, Caspase-8, ERCC1, Rb & Related to apoptosis\\ \hline \textbf{``Akt''} & Akt, Tuberin, STAT5A &\\ \end{longtable} \begin{figure}[!p] \fbox{ \begin{minipage}{6.5in} \includegraphics[width=\textwidth]{figure2.png} \caption{ Scatter plots connecting the LYM, M+ and M- metagenes in 12 cancer types. Each dot represents a cancer sample. The horizontal and vertical axes measure the average methylation values of the two methylation signatures, M- and M+, while the value of the expression of the LYM metagene is color-coded. In all three cases, the metagene is defined by the average of the top ranked genes as described in \textbf{Table ~\ref{tab:tabS1}}. } \label{fig:fig2} \end{minipage} } \end{figure} \subsection{Lymphocyte infiltration: LYM mRNA signature; M+\hspace{1pt} methylation signature; M-\hspace{1pt} methylation signature} These three signatures are related to tumor infiltration by lymphocytes. We list them together because they are strongly interrelated (\textbf{Fig. ~\ref{fig:fig2}}) even though each of the three was independently derived using an unsupervised computational method. The presence of LYM is accompanied by the presence of M+ and the absence of M- in all solid cancer types, suggesting that the three signatures reflect the same biomolecular event, which appears to be the infiltration of immune cells in tumor tissue. Indeed, there is remarkable similarity (\textbf{Fig. ~\ref{fig:fig3}}) between the LYM signature and the ``immune score'' of the ESTIMATE tumor purity computational tool (\url{http://ibl.mdanderson.org/estimate}). The values of the M+ methylation signature are also remarkably similar to those of the methylation-based ``leukocyte percentage'' estimation \cite{huiNature} (available under Synapse ID syn1809222). \begin{figure}[!p] \fbox{ \begin{minipage}{6.5in} \includegraphics[width=\textwidth]{figure3.png} \caption{ Scatter plots demonstrating the pan-cancer similarity of the value of the LYM metagene with the immune score of the ESTIMATE tumor purity computational tool (http://ibl.mdanderson.org/estimate) measuring immune cell infiltration. Each dot represents a cancer sample. The horizontal axis measures the expression value of the LYM metagene and the vertical axis measures the ESTIMATE immune score of infiltration. Note that the ESTIMATE did not provide scores for rectum cancer, and the estimation of immune cell infiltration is not applicable in leukemia. } \label{fig:fig3} \end{minipage} } \end{figure} We had previously found\cite{billCancerInfo} all three LYM, M+ and M- signatures from their association with the expression of miR-142. We have now confirmed this association with miR-142 in the pancan12 data sets, and we found that miR-150 and miR-155 are also strongly associated with the LYM signature. We had also previously independently identified the LYM signature as an attractor metagene\cite{mePLoS}, and used it in the winning model of the Sage Bionetworks Breast Cancer Prognosis Challenge\cite{meSTM}. Specifically the LYM signature is strongly associated with improved prognosis in ER-negative breast cancers, and this fact also provides an explanation for the relatively better prognosis in medullary, compared with other types of high-grade breast cancers. The interrelationship of the LYM, M+ and M- signatures, as shown in \textbf{Fig. ~\ref{fig:fig2}}, appears to be a consequence of the presence of different subclasses of cells (as opposed to being a methylation switch inside the same cell), consistent with their assumed role of measuring the extent of lymphocyte infiltration in the tumor. In other words, the M+ methylation sites, normally unmethylated, are largely methylated in the infiltrating leukocytes; and the M- methylation sites, normally methylated, are largely unmethylated in the infiltrating leukocytes. Consistently, many of the genes methylated by the M- signature are identical to those of LYM (six among the 27 genes of the M- signature (\textit{BIN2}, \textit{TNFAIP8L2}, \textit{ACAP1}, \textit{NCKAP1L}, \textit{FAM78A}, \textit{PTPN7}) listed in Table S1 are also among the 168 genes listed in the LYM attractor metagene ($P < 9.21\times10^{-7}$ based on Fisher’s exact test). The observed significant overlap in the gene sets and the negative association between gene expression in LYM and DNA methylation in M- are consistent with the notion that the absence of DNA methylation is permissive for gene expression, suggesting that the expression of the LYM signature in the infiltrating lymphocytes may be facilitated in part by the hypomethylation of the M- signature. The sharp definition of the LYM signature (being a Pan-Cancer attractor signature pointing to few genes at the core of coexpression) provides strong hints about the precise nature of this leukocyte infiltration. Specifically, the membership of the top-ranked genes (\textit{SASH3}, \textit{CD53}, \textit{NCKAP1L}, \textit{LCP2}, \textit{IL10RA}, \textit{PTPRC}, \textit{EVI2B}, \textit{BIN2}, \textit{WAS}, \textit{HAVCR2}, \ldots) point to a specific type of lymphocytes. We have speculated\cite{mePLoS} that these infiltrating lymphocytes are T cells having undergone a particular type of co-stimulation providing hypotheses for related adoptive transfer therapy. Two proteins strongly associated with the LYM signature are two tyrosine kinases: Lck (lymphocyte-specific protein tyrosine kinase) and Syk (spleen tyrosine kinase). \subsection{CIN (mitotic chromosomal instability) mRNA signature} This signature is related to mitotic chromosomal instability. It is similar to numerous known ``proliferation'' signatures, but its sharp definition as an attractor metagene specifically points to the kinetochore-microtubule interface and associated kinesins. Comparison with similar mitotic signatures in normal cells may help pinpoint driver genes for malignant chromosomal instability. The signature is strongly associated with tumor grade as well as poor prognosis in many, if not all, cancer types. Two proteins strongly associated with the CIN signature are Cyclin B1 and CDK1. Consistently, it is known that the cyclin B1-Cdk1 complex of cyclin-dependent kinase 1 is involved in the early events of mitosis, and that nuclear cyclin B1 protein may induce chromosomal instability and enhance the aggressiveness of the carcinoma cells\cite{bibi7}. \subsection{MES (mesenchymal transition) mRNA signature} This signature is related to mesenchymal transition and invasiveness of cancer cells. It is similar to numerous ``stromal'' or ``mesenchymal'' signatures; however there is evidence\cite{dimitrisBMCCancer} that many among the genes of the signature are largely produced by transdifferentiated cancer cells. We hypothesize that such cells, known to assume the duties of cancer-associated fibroblasts in some tumors\cite{hanahan2011}, may have become indistinguishable, even using laser capture microdissection, from stromal fibroblasts. We had originally identified the MES signature from its association with tumor stage\cite{hoonBMCMedGenomics}; specifically the signature appears only after a particular cancer type-specific tumor stage threshold has been reached. The values of the MES signature are remarkably similar to the ``stromal score'' of the ESTIMATE tumor purity computational tool (\url{http://ibl.mdanderson.org/estimate}) measuring fibroblast infiltration. Based on our previous reasoning, however, we believe that this interpretation may not be fully accurate, and that it will be important to find out to what extent some of the cells expressing some of these mesenchymal markers may actually be transdifferentiated cancer cells, and whether the estimated tumor purity may be affected by other types of normal cells instead of stromal fibroblasts. The co-regulated microRNAs most strongly associated with the MES signature are miR-199a, miR-199b, and miR-214. The \textit{DLK1}- \textit{DIO3} RNA cluster attractor signature, described later, is also strongly associated with MES. The protein most strongly associated with the MES signature is Fibronectin. \subsection{END (endothelial marker) mRNA signature} This is a novel angiogenesis-associated attractor signature. Nearly all the top-ranked genes (\textbf{Table 1}) are endothelial markers. The top gene, \textit{CDH5}, codes for VE-cadherin, which is known to be involved in a pathway suppressing angiogenic sprouting\cite{bib9}. The second gene, \textit{ROBO4}, is known to inhibit VEGF-induced pathologic angiogenesis and endothelial hyperpermeability\cite{bib10}. Consistently, the END attractor metagene appears to be protective and anti-angiogenic, stabilizing the vascular network. For example, 22 out of the 27 genes of the END attractor are among the 265 genes included in File S2 of a recent study\cite{bib11} of renal cell carcinoma ($P < 8.4\times10^{-38}$ based on Fisher’s exact test) as most associated with patients’ survival. These good-prognosis genes were intermixed in the same file with many poor-prognosis genes of the CIN attractor, suggesting that the CIN and END attractor metagenes are two of the most prognostic features in renal cell carcinoma. Interestingly, the MES and END attractor metagenes are positively associated with each other (\textbf{Fig. ~\ref{fig:fig4}}), in the sense that overexpression of the END signature tends to imply overexpression of the MES signature and vice-versa. This is consistent with mutual exclusivity between angiogenesis and invasiveness and with related findings\cite{bib12} that VEGF inhibits tumor cell invasion and mesenchymal transition, while antiangiogenic therapy is associated with increased invasiveness\cite{bib13}. It may also explain the paradoxical protective nature of signatures related to the MES attractor metagene in invasive breast cancers\cite{bib14}. \begin{figure}[!p] \fbox{ \begin{minipage}{6.5in} \includegraphics[width=\textwidth]{figure4.png} \caption{ Scatter plots demonstrating the association between MES and END attractor metagenes. The horizontal and vertical axes measure the values of the MES and END signatures. The two signatures have positive correlation, although this association is not sufficiently strong to merge the two attractors into one. This association suggests that the invasive MES signature and the antiangiogenic END signature tend to be present simultaneously. } \label{fig:fig4} \end{minipage} } \end{figure} \subsection{``AHSA2'' mRNA signature} We do not yet know what this signature represents. We observed that several noncoding RNAs (e.g. NCRNA00105, NCRNA00201) are in relatively high-ranked positions among its members. \subsection{IFIT (interferon-induced) mRNA signature} The members of this signature are interferon-induced. For example, we observed large enrichment of the genes of the signature among those upregulated by IFN-$\alpha$ in the side population (SP) of ovarian cancer cells\cite{bib15} from the list provided in Supplementary Table S4 of that paper, in which the authors concluded that tumors bearing large SP numbers could be particularly sensitive to IFN-$\alpha$ treatment. \subsection{``WDR38'' mRNA signature} We do not know what this signature represents, except that we had found one of its key members, gene \textit{ZMYND10}, to be protective and associated with estrogen receptor expression in breast cancer. \subsection{MHC Class II genomically co-localized mRNA signature} We found this signature using the genomically co-localized version of the algorithm. It is very highly correlated with LYM. \subsection{GIMAP genomically co-localized mRNA signature} As above, we found this signature using the genomically co-localized version of the algorithm. It is also very highly correlated with LYM. \subsection{Chr8q24.3 amplicon mRNA signature} This is the strongest pan-cancer amplicon signature. It was previously found predictive of early relapse in ER-positive breast cancers\cite{bib16}. \subsection{``RMND1'' methylation signature} We do not yet know what the comethylation of the sites of this signature signifies. \begin{figure}[!t] \fbox{ \begin{minipage}{6.5in} \includegraphics[width=\textwidth]{figure5.png} \caption{ The \textit{DLK1}-\textit{DIO3} cluster of noncoding RNAs. Shown is a screen capture from the UCSC Genome Browser (\url{http://genome.ucsc.edu}). The cluster of imprinted genes delineated by the \textit{DLK1} and \textit{DIO3} genes (outside the shown region) is located on chromosome 14. We found that the corresponding pan-cancer attractor signature does not contain any paternally inherited protein-coding genes. It does contain the numerous noncoding RNA genes expressed from the maternally inherited homolog, including the \textit{MEG3} long noncoding RNA gene. } \label{fig:fig5} \end{minipage} } \end{figure} \subsection{\textit{DLK1}-\textit{DIO3} RNA cluster signature} This is the strongest pan-cancer multi-microRNA coexpression signature. It consists of numerous noncoding RNAs within the \textit{DLK1}-\textit{DIO3} imprinted genomic region of chr14q32. \textbf{Fig. ~\ref{fig:fig5}} shows a screen capture of the genomic region from the UCSC Genome Browser (\url{http://genome.ucsc.edu/}). We confirmed that the coexpression signature also includes the \textit{MEG3} long noncoding RNA located at the upstream end of the region. It may also include numerous small nuclear RNAs at the central region, but there were no associated probe sets to confirm the coexpression. We found that this ncRNA signature is associated with the MES (mesenchymal transition) mRNA signature. For example, the ranked list of mRNAs most associated with the \textit{DLK1}-\textit{DIO3} ncRNA signature starts from \textit{POSTN}, \textit{PCOLCE}, \textit{COL5A2}, \textit{COL1A2}, \textit{GLT8D2}, \textit{COL5A1}, \textit{SFRP2}, and \textit{FAP}. Expression of the imprinted \textit{DLK1}-\textit{DIO3} ncRNA cluster is believed to be vital for the development potential of embryonic stem cells\cite{bib17}, consistent with the hypothesis\cite{bib18} that mesenchymal transition in cancer reactivates embryonic developmental programs and makes cancer cells invasive and stem-like. The \textit{DLK1}-\textit{DIO3} ncRNA signature was also found to define a stem-like subtype of hepatocellular carcinoma associated with poor survival\cite{bib19}. The details of the regulation mechanism for this ncRNA cluster coexpression in the \textit{DLK1}-\textit{DIO3} region are unclear. \subsection{``miR-509/miR-514/miR-508'' microRNA signature} These three microRNAs are co-localized at chrXq27.3. We do not know what this signature signifies. \subsection{``miR-144/miR-451/miR-486'' microRNA signature} This is a three-microRNA signature related to erythropoiesis. The first two genes are located in the bicistronic microRNA locus miR-144/451, highly expressed during erythrocyte development\cite{bib20}. The mRNAs most associated to this microRNA signature are hemoglobin-related: \textit{HBB}, \textit{HBA1}, \textit{HBA2} and \textit{ALAS2}. The protein most associated with this signature is HER3. These three microRNAs were identified as promising biomarkers for detection of esophageal cancer. \subsection{c-Met/Snail/PARP\_cleaved/Caspase-8/ERCC1/Rb protein activity signature} This protein coexpression signature appears to combine the contribution of several pathways and we hope that a plausible and useful biological ``story'' will be developed based on the simultaneous activity of all these six proteins in some cancer samples. We note that each of these proteins\cite{bib21, bib22, bib23, bib24, bib25, bib26} has been related in various ways with resistance to chemotherapy or apoptosis. \subsection{Akt/Tuberin/STAT5A protein activity signature} We do not know what the coexpression of Akt, Tuberin, STAT5A proteins represents in cancer. It is known, however, that low levels of STAT5A protein in breast cancer are associated with tumor progression and unfavorable clinical outcomes\cite{bib27}. \section*{DISCUSSION} The Pan-Cancer nature (\textbf{Fig. ~\ref{fig:figS1}}) of each of the signatures described in this paper suggests that they represent important biomolecular events. A reasonable concern is whether some of these ``pan-cancer'' signatures may instead reflect fundamental normal ``pan-tissue'' biological mechanisms. Even if this is true for some of these signatures, this does not exclude the possibility that they are aberrant and play important roles in some cancer samples. Furthermore, this provides the opportunity to compare similar signatures in normal vs. malignant tissues to pinpoint potential cancer-specific genes. Because of its exhaustive search starting from all potential ``seeds'' in all data sets from twelve different cancer types, our iterative data mining algorithm is guaranteed to have identified all pan-cancer molecular signatures involving simultaneous presence of a large number of coordinately expressed genes, proteins, or comethylated sites. We hope that these signatures are further scrutinized by the medical research community for the purpose of developing potential diagnostic, predictive, and eventually therapeutic products applicable in multiple cancers. \section*{ACKNOWLEDGEMENTS} We are thankful to Hanina Hibshoosh, Chris Miller and Gordon Mills for helpful discussions, which contributed to improved interpretation of the signatures disclosed in this paper. \section*{Accessibility} All figures in this paper, including supplementary figures and tables, as well as the files of generated attractor molecular signatures, are available in Synapse under ID syn1686966. \section*{Data description and availability} The data sets of TCGA pancan12 freeze 4.7 used to derive the results of this paper are described and are available under Synapse ID syn300013 with doi:10.7303/syn300013. The twelve cancer types are bladder urothelial carcinoma (BLCA), breast invasive carcinoma (BRCA), colon adenocarcinoma (COAD), glioblastoma multiforme (GBM), head and neck squamous cell carcinoma (HNSC), kidney renal clear cell carcinoma (KIRC), acute myeloid leukemia (LAML), lung adenocarcinoma (LUAD) , lung squamous cell carcinoma (LUSC), ovarian serous cystadenocarcinoma (OV), rectum adenocarcinoma (READ), and uterine corpus endometrioid carcinoma (UCEC). \pagebreak %% Put the bibliography here, most people will use BiBTeX in %% which case the environment below should be replaced with %% the \bibliography{} command. \begin{thebibliography}{1} \bibitem{hanahan2000} Hanahan, D. \& Weinberg, R.A. The hallmarks of cancer. Cell 100, 57-70 (2000). \bibitem{hanahan2011} Hanahan, D. \& Weinberg, R.A. Hallmarks of cancer: the next generation. Cell 144, 646-74 (2011). \bibitem{mePLoS} Cheng, W.Y., Ou Yang, T.H. \& Anastassiou, D. Biomolecular events in cancer revealed by attractor metagenes. PLoS Comput Biol 9, e1002920 (2013). \bibitem{meSTM} Cheng, W.Y., Ou Yang, T.H. \& Anastassiou, D. Development of a prognostic model for breast cancer survival in an open challenge environment. Sci Transl Med 5, 181ra50 (2013). \bibitem{bib5} McCarthy, N. Prognostic models: Rising to the challenge. Nat Rev Cancer, doi:10.1038/nrc3530 (2013). \bibitem{huiNature} Shen, H. \textit{et al.} Comprehensive Cross-Cancer Comparison of DNA Methylation Profiles, submitted. (2013). \bibitem{billCancerInfo} Andreopoulos, B. \& Anastassiou, D. Integrated Analysis Reveals hsa-miR-142 as a Representative of a Lymphocyte-Specific Gene Expression and Methylation Signature. Cancer Inform 11, 61-75 (2012). \bibitem{bibi7} Suzuki, T. \textit{et al.} Nuclear cyclin B1 in human breast carcinoma as a potent prognostic factor. Cancer Sci 98, 644-51 (2007). \bibitem{dimitrisBMCCancer} Anastassiou, D. \textit{et al.} Human cancer cells express Slug-based epithelial-mesenchymal transition gene expression signature obtained in vivo. BMC Cancer 11, 529 (2011). \bibitem{hoonBMCMedGenomics} Kim, H., Watkinson, J., Varadan, V. \& Anastassiou, D. Multi-cancer computational analysis reveals invasion-associated variant of desmoplastic reaction involving INHBA, THBS2 and COL11A1. BMC Med Genomics 3, 51 (2010). \bibitem{bib9} Abraham, S. \textit{et al.} VE-Cadherin-mediated cell-cell interaction suppresses sprouting via signaling to MLC2 phosphorylation. Curr Biol 19, 668-74 (2009). \bibitem{bib10} Jones, C.A. \textit{et al.} Robo4 stabilizes the vascular network by inhibiting pathologic angiogenesis and endothelial hyperpermeability. Nat Med 14, 448-53 (2008). \bibitem{bib11} Wozniak, M.B. \textit{et al.} Integrative genome-wide gene expression profiling of clear cell renal cell carcinoma in Czech Republic and in the United States. PLoS One 8, e57886 (2013). \bibitem{bib12} Lu, K.V. \textit{et al.} VEGF inhibits tumor cell invasion and mesenchymal transition through a MET/VEGFR2 complex. Cancer Cell 22, 21-35 (2012). \bibitem{bib13} Paez-Ribes, M. \textit{et al}. Antiangiogenic therapy elicits malignant progression of tumors to increased local invasion and distant metastasis. Cancer Cell 15, 220-31 (2009). \bibitem{bib14} Beck, A.H., Espinosa, I., Gilks, C.B., van de Rijn, M. \& West, R.B. The fibromatosis signature defines a robust stromal response in breast carcinoma. Lab Invest 88, 591-601 (2008). \bibitem{bib15} Moserle, L. \textit{et al}. The side population of ovarian cancer cells is a primary target of IFN-alpha antitumor effects. Cancer Res 68, 5658-68 (2008). \bibitem{bib16} Bilal, E. \textit{et al}. Amplified Loci on Chromosomes 8 and 17 Predict Early Relapse in ER-Positive Breast Cancers. PLoS One 7, e38575 (2012). \bibitem{bib17} Stadtfeld, M. \textit{et al}. Aberrant silencing of imprinted genes on chromosome 12qF1 in mouse induced pluripotent stem cells. Nature 465, 175-81 (2010). \bibitem{bib18} Mani, S.A. \textit{et al}. The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell 133, 704-15 (2008). \bibitem{bib19} Luk, J.M. \textit{et al}. DLK1-DIO3 genomic imprinted microRNA cluster at 14q32.2 defines a stemlike subtype of hepatocellular carcinoma associated with poor survival. J Biol Chem 286, 30706-13 (2011). \bibitem{bib20} Yu, D. \textit{et al}. miR-451 protects against erythroid oxidant stress by repressing 14-3-3zeta. Genes Dev 24, 1620-33 (2010). \bibitem{bib21} Tang, M.K., Zhou, H.Y., Yam, J.W. \& Wong, A.S. c-Met overexpression contributes to the acquired apoptotic resistance of nonadherent ovarian cancer cells through a cross talk mediated by phosphatidylinositol 3-kinase and extracellular signal-regulated kinase 1/2. Neoplasia 12, 128-38 (2010). \bibitem{bib22} Haslehurst, A.M. \textit{et al}. EMT transcription factors snail and slug directly contribute to cisplatin resistance in ovarian cancer. BMC Cancer 12, 91 (2012). \bibitem{bib23} D’Amours, D., Sallmann, F.R., Dixit, V.M. \& Poirier, G.G. Gain-of-function of poly(ADP-ribose) polymerase-1 upon cleavage by apoptotic proteases: implications for apoptosis. J Cell Sci 114, 3771-8 (2001). \bibitem{bib24} Kim, P.K., Mahidhara, R. \& Seol, D.W. The role of caspase-8 in resistance to cancer chemotherapy. Drug Resist Updat 4, 293-6 (2001). \bibitem{bib25} Olaussen, K.A. \textit{et al}. DNA repair by ERCC1 in non-small-cell lung cancer and cisplatin-based adjuvant chemotherapy. N Engl J Med 355, 983-91 (2006). \bibitem{bib26} Collard, T.J. \textit{et al}. The retinoblastoma protein (Rb) as an anti-apoptotic factor: expression of Rb is required for the anti-apoptotic function of BAG-1 protein in colorectal tumour cells. Cell Death Dis 3, e408 (2012). \bibitem{bib27} Peck, A.R. \textit{et al}. Low levels of Stat5a protein in breast cancer are associated with tumor progression and unfavorable clinical outcomes. Breast Cancer Res 14, R130 (2012). \end{thebibliography} \pagebreak \begin{methods} \subsection{Data normalization} The data platform for each cancer types and its corresponding Synapse ID is given below. \begin{table}[!hf] \renewcommand{\arraystretch}{1} \hspace*{-0.5in} \begin{threeparttable} \begin{tabular}{ |l|c|c|c|c|} \hline \textbf{Molecular profile} & mRNA & Protein & miRNA & DNA methylation \\ \hline \multirow{3}{*}{\textbf{Platform}} & & Reverse phase protein & & Infinium \\ & Illumina HiSeq & lysate microarray & Illumina HiSeq & HumanMethylation27 \\ & & (RPPA) & & BeadChip \\ \hline \textbf{Cancer type} & \multicolumn{4}{|c|}{\textbf{Synapse ID}} \\ \hline BLCA&syn1571504&syn1681048&syn1571494&syn1889358\tnote{*}\\ BRCA&syn417812&syn1571267&syn395575&syn411485\\ COAD&syn1446197&syn416772&syn464211&syn411993\\ GBM&syn1446214&syn416777&NA&syn412284\\ HNSC&syn1571420&syn1571409&syn1571411&syn1889356\tnote{*}\\ KIRC&syn417925&syn416783&syn395617&syn412701\\ LAML&syn1681084&NA&syn1571533&syn1571536\\ LUAD&syn1571468&syn1571446&syn1571453&syn1571458\\ LUSC&syn418033&syn1367036&syn395691&syn415758\\ OV&syn1446264&syn416789&syn1356544&syn415945\\ READ&syn1446276&syn416795&syn464222&syn416194\\ UCEC&syn1446289&syn416800&syn395720&syn416204\\ \hline \end{tabular} \begin{tablenotes} \item[*] The data sets were extracted from HumanMethylation450 BeadChip. \end{tablenotes} \end{threeparttable} \end{table}

2304.00477v2

\begin{table}[h] \centering % \vspace{-5pt} \begin{threeparttable} \resizebox{\linewidth}{!}{ \begin{tabular}{|l|c|c|c|} \hline & \tool & Code Interpreter & Pandas Agent \\ \hline \textbf{Relevance} & {\bf 4.50$\pm$0.76} & 4.08$\pm$0.86 & 1.92$\pm$1.00 \\ \hline \textbf{Completeness} & {\bf 4.67$\pm$0.55} & 3.54$\pm$1.00 & 1.12$\pm$0.33 \\ \hline \textbf{Understandability} & {\bf 4.46$\pm$0.64} & 4.25$\pm$0.83 & 1.62$\pm$0.90 \\ \hline \end{tabular}} % \vspace{-10pt} \caption{Results of the User Study} \label{tab:user_study} \end{threeparttable} % \vspace{-10pt} \end{table}

1904.03259v1

\begin{table} \begin{small} \begin{center} \begin{tabular}{|c|c|} \hline \textbf{Algorithm} & \textbf{Supervision} \\ \hline\hline Kmeans clustering & Internal\\ Variational autoencoders & Internal\\ Deep belief networks & Internal\\ Principal component analysis & Internal\\ K nearest neighbor (kNN) & External\\ CNN image classification & External\\ Generative adversarial nets & External\\ Kmeans-to-kNN & Both\\ \hline \end{tabular} \end{center} \end{small} \caption[category]{Supervision type of some algorithms} \label{tab:category} \end{table}

1910.12734v1

\begin{table}[htbp] \small \centering \begin{tabular}{|l|p{1.5cm}|p{5cm}|} \hline Date & Place & Description \\ \hline 5/30/06 & Palazzo del Quirinale & On. Sen. Franco MARINI, Presidente del Senato della Repubblica, e On. Fausto BERTINOTTI, Presidente della Camera dei Deputati \\ \hline 6/7/06 & Palazzo del Quirinale & On. Silvio BERLUSCONI, Presidente di Forza Italia \\ \hline 6/3/08 & Palazzo della FAO & Intervento alla cerimonia di apertura della Conferenza sulla,sicurezza alimentare, promossa dalla FAO \\ \hline \end{tabular} \caption{Sample records from the Diary of the President of the Italian Republic, Giorgio Napolitano.} \label{tab:sample_diary} \vspace{-20pt} \end{table}

2405.13998v1

\begin{table}[h] \centering \renewcommand{\arraystretch}{1.2} \caption{Summary of the main symbols and notation used in this work.} \begin{tabular}{|>{\centering\arraybackslash}m{0.2\textwidth}|>{\arraybackslash}m{0.7\textwidth}|} \hline \rowcolor{gray!30} \textbf{Notation} & \textbf{Description} \\ \hline \multicolumn{2}{|>{\columncolor{gray!15}}c|}{\textbf{Operator Learning}} \\ \hline $\mathcal{X}$ & The input function space \\ $\mathcal{Y}$ & The output function space \\ $u \in \mathcal{X}$ & Input function \\ $s \in \mathcal{Y}$ & Output function \\ $y$ & Query coordinate in the input domain of $s$ \\ $\mathcal{G}: \mathcal{X} \rightarrow \mathcal{Y}$ & The operator mapping between function spaces \\ $\curlyE: \curlyX \to \R^n$ & Encoder mapping \\ $\curlyD: \R^n \to \curlyY $ & Decoder mapping \\ \hline \multicolumn{2}{|>{\columncolor{gray!15}}c|}{\textbf{Fourier Neural Operator}} \\ \hline $\mathcal{F}, \mathcal{F}^{-1}$ & Fourier transform and its inverse \\ $\mathcal{F}_n, \mathcal{F}_n^{-1}$ & Discrete Fourier transform and its inverse truncated on the first $n$ modes \\ $K$ & Linear transformation applied to the $n$ leading Fourier modes \\ $W$ & Linear transformation (bias term) applied to the layer inputs \\ $k$ & Fourier modes / wave numbers \\ % $k_{\text{max}}$ & Max Fourier modes used in the Fourier layer \\ $\sigma$ & Activation function \\ \hline \multicolumn{2}{|>{\columncolor{gray!15}}c|}{\textbf{Continuous Vision Transformer}} \\ \hline $\text{PE}_{t}$ & Temporal positional embedding \\ $\text{PE}_{s}$ & Spatial positional embedding \\ % $\text{SA}$ & Self-attention \\ $\text{MSA}$ & Multi-head self-attention \\ $\text{MHA}$ & Multi-head attention \\ % $\text{CSA}$ & Multi-head cross-attention \\ $\text{LN}$ & Layer normalization \\ $P$ & Patch size of Vision Transformer \\ $C$ & Embedding dimension of Vision Transformer \\ % $N_x \times N_y$ & Resolution of dummy grid \\ $\mathbf{x} \in \mathbb{R}^{N_x \times N_y \times C}$ & Latent grid features \\ $\epsilon$ & Locality of interpolated latent grid features \\ \hline \multicolumn{2}{|>{\columncolor{gray!15}}c|}{\textbf{Partial Differential Equations}} \\ \hline $\zeta$ & Vorticity of shallow water equations \\ $\eta$ & Height of shallow water equations \\ $\mathbf{u}$ & Velocity field \\ $P$ & Pressure field \\ $c$ & Passive scalar (smoke) \\ % $\mathbf{f}$ & Buoyancy \\ $\nu$ & Viscosity \\ \hline \multicolumn{2}{|>{\columncolor{gray!15}}c|}{\textbf{Hyperparameters}} \\ \hline $B$ & Batch size \\ $Q$ & Number of query coordinates in each batch \\ $D$ & Number of latent variables of interest \\ $T$ & Number of previous time-steps \\ $H \times W$ & Resolution of spatial discretization \\ \hline \end{tabular} \label{tab:notation} \end{table}

1809.07879v1

\begin{table}\label{dataset} \centering \begin{threeparttable} {\footnotesize \caption{Data set} \begin{tabular}[H]{|c|c|c|c|c|c|c|c|c|c|} \hline Star & P & T &H&rms1&img&rms2&A&z&V\\ &(HPa)&($^{\circ}{\rm C}$)&(\%)&(px)&&(px)&($^{\circ}...$)&($^{\circ}...$)&($^{\circ}$/h)\\ \hline \color{gray}1&\color{gray}2&\color{gray}3&\color{gray}4&\color{gray}5&\color{gray}6&\color{gray}7&\color{gray}8&\color{gray}9&\color{gray}10\\ \hline $\alpha Cas$&1002.1&8.6&50.0&0.15&161&0.2&322.28286&60.91284 &8.27\\ \hline $\alpha Ori$ &1002.3&8.0&50.0&0.21&312&0.16&231.03903&48.43744&14.87 \\ \hline $\gamma Cep$& 1002.3&7.4&50.0 &0.21&263&0.23&350.05359&55.17984& 3.21\\ \hline $\alpha Hya$& 1002.4&7.1&50.0 &0.24&523&0.19&176.01978&53.27512&14.82\\ \hline \end{tabular}} \begin{tablenotes} \small \item 1 - Observed star, 2 - Pressure, 3 - Temperature, 4 - Humidity, 5 -root mean square 1 in pixels, 6 - number of images taken by the CCD camera, 7 - root mean square 2 in pixels, 8 - Azimuth, 9 - Zenith distance, 10 - Star apparent velocity \end{tablenotes} \end{threeparttable} \end{table}

1809.07879v1

\begin{table} \centering \caption{Percentage of inertia for each dimension and cumulative percentage of inertia} {\footnotesize \begin{tabular}[!h]{|c|c|c|} \hline Dimension&Percentage of inertia&Cumulative percentage of inertia\\ \hline 1&$23.04\%$&$23.04\%$ \\ \hline 2&$18.62\%$&$41.66\%$\\ \hline 3&$16.56\%$&$58.23\%$\\ \hline 4&$12.74\%$&$70.98\%$\\ \hline 5&$9.53\%$&$80.51\%$\\ \hline 6&$7.25\%$&$87.77\%$\\ \hline 7&$5.36\%$&$93.14\%$\\ \hline 8&$3.74\%$&$96.89\%$\\ \hline 9&$3.11\%$&$100\%$\\ \hline \end{tabular} } \label{variance} \end{table}

1809.07879v1

\begin{table} \centering {\footnotesize \caption{Variables coorrelations with the first four dimesnions} \label{coordvar} \begin{tabular}[H]{|c|c|c|c|c|} \hline Variable&Dim.1&Dim.2&Dim.3&Dim.4\\ \hline P&0.22&\cellcolor{green!15}0.61&\cellcolor{green!15}0.51&-0.35\\ \hline T&\cellcolor{green!15}0.52&-0.06&-0.35&\cellcolor{green!15}-0.65\\ \hline H&0.03&0.38&\cellcolor{green!15}0.83&0.04\\ \hline rms1&\cellcolor{green!15}0.67&-0.44&0.39&0.03\\ \hline img&\cellcolor{green!15}-0.77&-0.23&0.20&0.07\\ \hline rms2&0.38&\cellcolor{green!15}-0.71&0.32&0.22\\ \hline A&0.11&\cellcolor{green!15}0.55&-0.25&0.46\\ \hline Z&0.37&0.08&-0.04&\cellcolor{green!15}0.56\\ \hline velocity&\cellcolor{green!15}0.64&0.29&-0.22&0.16\\ \hline \end{tabular}} \end{table}

2007.15745v3

\begin{table} \centering \caption{Configuration space for the hyper-parameters of tested ML models} \setlength\extrarowheight{1pt} \centering \scriptsize \begin{tabular}{l|l|l|l} \Xhline{1.2pt} \textbf{ML Model } & \multicolumn{1}{l|}{\textbf{Hyper-parameter}} & \multicolumn{1}{l|}{\textbf{Type}} & \textbf{Search Space} \\ \Xhline{1.2pt} \multirow{6}{*}{RF Classifier} & n\_estimators & Discrete & {[}10,100] \\ \cline{2-4} & max\_depth & \multicolumn{1}{l|}{Discrete} & {[}5,50] \\ \cline{2-4} & min\_samples\_split & Discrete & {[}2,11] \\ \cline{2-4} & min\_samples\_leaf & Discrete & {[}1,11] \\ \cline{2-4} & criterion & \multicolumn{1}{l|}{Categorical} & {[}'gini', 'entropy'] \\ \cline{2-4} & max\_features & Discrete & {[}1,64] \\ \hline \multirow{2}{*}{SVM Classifier} & C & Continuous & {[}0.1,50] \\ \cline{2-4} & kernel & Categorical & {[}'linear', 'poly', 'rbf', 'sigmoid'] \\ \hline KNN Classifier & n\_neighbors & Discrete & {[}1,20] \\ \hline \multirow{6}{*}{RF Regressor} & n\_estimators & Discrete & {[}10,100] \\ \cline{2-4} & max\_depth & Discrete & {[}5,50] \\ \cline{2-4} & min\_samples\_split & Discrete & {[}2,11] \\ \cline{2-4} & min\_samples\_leaf & Discrete & {[}1,11] \\ \cline{2-3}\cline{4-4} & criterion & Categorical & {[}'mse', 'mae'] \\ \cline{2-4} & max\_features & Discrete & {[}1,13] \\ \hline \multirow{3}{*}{SVM Regressor} & C & Continuous & {[}0.1,50] \\ \cline{2-4} & kernel & Categorical & {[}'linear', 'poly', 'rbf', 'sigmoid'] \\ \cline{2-4} & epsilon & Continuous & {[}0.001,1] \\ \hline KNN Regressor & n\_neighbors & Discrete & {[}1,20] \\ \Xhline{1.2pt} \end{tabular} \label{ta1}% \end{table}

2007.15745v3

\begin{table} \centering \caption{Performance evaluation of applying HPO methods to the RF classifier on the MNIST dataset} \setlength\extrarowheight{1pt} \centering \scriptsize \begin{tabular}{p{2cm}|p{1.5cm}|p{1.5cm}} \Xhline{1.2pt} \textbf{Optimization Algorithm} & \textbf{Accuracy} (\%) & \textbf{CT} (s) \\ \Xhline{1.2pt} Default HPs & 90.65 & 0.09 \\ \hline GS & 93.32 & 48.62 \\ \hline RS & 93.38 & 16.73 \\ \hline BO-GP & 93.38 & 20.60 \\ \hline BO-TPE & 93.88 & 12.58 \\ \hline Hyperband & 93.38 & 8.89 \\ \hline BOHB & 93.38 & 9.45 \\ \hline GA & 93.83 & 19.19 \\ \hline PSO & 93.73 & 12.43 \\ \Xhline{1.2pt} \end{tabular} \label{te1}% \end{table}

2007.15745v3

\begin{table} \centering \caption{Performance evaluation of applying HPO methods to the SVM classifier on the MNIST dataset} \setlength\extrarowheight{1pt} \centering \scriptsize \begin{tabular}{p{2cm}|p{1.5cm}|p{1.5cm}} \Xhline{1.2pt} \textbf{Optimization Algorithm} & \textbf{Accuracy} (\%) & \textbf{CT} (s) \\ \Xhline{1.2pt} Default HPs & 97.05 & 0.29 \\ \hline GS & 97.44 & 32.90 \\ \hline RS & 97.35 & 12.48 \\ \hline BO-GP & 97.50 & 17.56 \\ \hline BO-TPE & 97.44 & 3.02 \\ \hline Hyperband & 97.44 & 11.37 \\ \hline BOHB & 97.44 & 8.18 \\ \hline GA & 97.44 & 16.89 \\ \hline PSO & 97.44 & 8.33 \\ \Xhline{1.2pt} \end{tabular} \label{te2}% \end{table}

2007.15745v3

\begin{table} \centering \caption{Performance evaluation of applying HPO methods to the KNN classifier on the MNIST dataset} \setlength\extrarowheight{1pt} \centering \scriptsize \begin{tabular}{p{2cm}|p{1.5cm}|p{1.5cm}} \Xhline{1.2pt} \textbf{Optimization Algorithm} & \textbf{Accuracy} (\%) & \textbf{CT} (s) \\ \Xhline{1.2pt} Default HPs & 96.27 & 0.24 \\ \hline GS & 96.22 & 7.86 \\ \hline RS & 96.33 & 6.44 \\ \hline BO-GP & 96.83 & 1.12 \\ \hline BO-TPE & 96.83 & 2.33 \\ \hline Hyperband & 96.22 & 4.54 \\ \hline BOHB & 97.44 & 3.84 \\ \hline GA & 96.83 & 2.34 \\ \hline PSO & 96.83 & 1.73 \\ \hline \Xhline{1.2pt} \end{tabular} \label{te3}% \end{table}

2007.15745v3

\begin{table} \centering \caption{Performance evaluation of applying HPO methods to the RF regressor on the Boston-housing dataset} \setlength\extrarowheight{1pt} \centering \scriptsize \begin{tabular}{p{2cm}|p{1.5cm}|p{1.5cm}} \Xhline{1.2pt} \textbf{Optimization Algorithm} & \textbf{MSE} & \textbf{CT} (s) \\ \Xhline{1.2pt} Default HPs & 31.26 & 0.08 \\ \hline GS & 29.02 & 4.64 \\ \hline RS & 27.92 & 3.42 \\ \hline BO-GP & 26.79 & 17.94 \\ \hline BO-TPE & 25.42 & 1.53 \\ \hline Hyperband & 26.14 & 2.56 \\ \hline BOHB & 25.56 & 1.88 \\ \hline GA & 26.95 & 4.73 \\ \hline PSO & 25.69 & 3.20 \\ \Xhline{1.2pt} \end{tabular} \label{te4}% \end{table}

2007.15745v3

\begin{table} \centering \caption{Performance evaluation of applying HPO methods to the SVM regressor on the Boston-housing dataset} \setlength\extrarowheight{1pt} \centering \scriptsize \begin{tabular}{p{2cm}|p{1.5cm}|p{1.5cm}} \Xhline{1.2pt} \textbf{Optimization Algorithm} & \textbf{MSE} & \textbf{CT} (s) \\ \Xhline{1.2pt} Default HPs & 77.43 & 0.02 \\ \hline GS & 67.07 & 1.33 \\ \hline RS & 61.40 & 0.48 \\ \hline BO-GP & 61.27 & 5.87 \\ \hline BO-TPE & 59.40 & 0.33 \\ \hline Hyperband & 73.44 & 0.32 \\ \hline BOHB & 59.67 & 0.31 \\ \hline GA & 60.17 & 1.12 \\ \hline PSO & 58.72 & 0.53 \\ \Xhline{1.2pt} \end{tabular} \label{te5}% \end{table}

2007.15745v3

\begin{table} \centering \caption{Performance evaluation of applying HPO methods to the KNN regressor on the Boston-housing dataset} \setlength\extrarowheight{1pt} \centering \scriptsize \begin{tabular}{p{2cm}|p{1.5cm}|p{1.5cm}} \Xhline{1.2pt} \textbf{Optimization Algorithm} & \textbf{MSE} & \textbf{CT} (s) \\ \Xhline{1.2pt} Default HPs & 81.48 & 0.004 \\ \hline GS & 81.53 & 0.12 \\ \hline RS & 80.77 & 0.11 \\ \hline BO-GP & 80.77 & 0.49 \\ \hline BO-TPE & 80.83 & 0.08 \\ \hline Hyperband & 80.87 & 0.10 \\ \hline BOHB & 80.77 & 0.09 \\ \hline GA & 80.77 & 0.33 \\ \hline PSO & 80.74 & 0.19 \\ \Xhline{1.2pt} \end{tabular} \label{te6}% \end{table}

2209.12885v2

\begin{table}[ht] \caption{Hyperparameters and their chosen values} \centering \begin{tabular}{c|c} \midrule Hyperparameter & Value \\ \midrule Number of hidden layers & $3$ \\ Hidden layer size & $50$ \\ Step factor & $5$ \\ Training data size & $3\cdot 10^4$\\ \midrule \end{tabular} \label{tab:hparam_values} \end{table}

2209.12885v2

\begin{table}[ht] \caption{Hyperparameters and their possible values.} \centering \begin{tabular}{c|c} \midrule \\ Hyperparameter & Search space \\ \midrule Number of hidden layers & $\{2, 3, 4\}$ \\ Hidden layer size & $\{20, 30, 40, 50, 60, 70\}$ \\ Step factor & $\{5, 10, 15, 20, 25, 30\}$ \\ Training data size & $\{10^4, 2\cdot10^4, 3\cdot10^4, 4\cdot10^4, 5\cdot10^4, 6\cdot10^4\}$\\ \midrule \end{tabular} \label{tab:hparam} \end{table}

1307.0781v1

\begin{table}[t] \centering {\fontsize{8}{6}\selectfont \setlength{\tabcolsep}{.1em} \begin{tabular}{|l|c|c|c|c|c|} \hline Learner & 1 &2 &3 & 4\\ \hline Classification & Naive Bayes, &Always $1$, & RBF Network, & Random Tree, \\ Function (S1) & Logistic & Voted Perceptron &J48 & Always $0$\\ \hline Error & 47, & 53, & 47, & 47, \\ percentage (S1) & 3 & 4 & 47 & 47\\ \hline Classification & Naive Bayes, &Always $1$, & RBF Network, & Random Tree, \\ Function (S2) & Random & Random & J48 & Always $0$ \\ \hline Error & 47, & 53, & 47, & 47, \\ percentage (S2) & 50 & 50 & 47 & 47 \\ \hline \end{tabular} } \add{\vspace{-0.1in}} \caption{Simulation setup} \vspace{-0.25in} \label{tab:sim_setup} \end{table}

1307.0781v1

\begin{table}[t] \centering {\fontsize{8}{6}\selectfont \setlength{\tabcolsep}{.3em} \begin{tabular}{|l|c|c|c|c|} \hline & $D_1(t)$ & $D_2(t)$ & $D_3(t)$ & $m_T$ \\ \hline (C1) CoS & $t^{1/8} \log t$ & $2 t^{1/8} \log t$ & $t^{1/8} \log t$ & $\lceil T \rceil^{1/4}$ \\ \hline %(C1) DCZA & $t^{1/8} \log t$ & $2 t^{1/8} \log t$ & $t^{1/8} \log t$ & & $1$ & $4$ \\ %\hline (C2) CoS & $t^{1/2} \log t$ & $2 t^{1/2} \log t$ & $t^{1/2} \log t$ & $\lceil T \rceil^{1/4}$ \\ \hline %(C2) DCZA & $t^{2/p} \log t$ & $2 t^{2/p} \log t$ & $t^{2/p} \log t$ & & $1$ & $(3+\sqrt{17})/2$ \\ %\hline \end{tabular} } \add{\vspace{-0.05in}} \caption{Parameters for CoS} \label{tab:par_setup} \add{\vspace{-0.4in}} \end{table}

1307.0781v1

\begin{table}[t] \centering {\renewcommand{\arraystretch}{0.6} {\fontsize{8}{6}\selectfont \setlength{\tabcolsep}{.1em} \begin{tabular}{|l|c|c|c|c|c|} \hline & \cite{breiman1996bagging, buhlmann2003boosting, lazarevic2001distributed, chen2004channel, perlich2011cross} & \cite{mateos2010distributed, kargupta1999collective} & \cite{zheng2011attribute} & This work \\ \hline Aggregation & non-cooperative & cooperative & cooperative & \rev{no} \\ \hline Message & none & data & training & data and label \\ exchange & & & residual & only if improves \\ & & & & performance \\ \hline Learning & offline/online & offline & offline & Non-bayesian \\ approach&&&& online\\ \hline Correlation & N/A & no & no & yes\\ exploitation & & & &\\ \hline Information from & no & all & all & only if improves \\ other learners & & & & accuracy \\ \hline Data partition & horizontal & horizontal & vertical & horizontal \\ \hline Bound on regret, & no &no &no &yes - sublinear\\ convergence rate &&&&\\ \hline \end{tabular} } } \caption{Comparison with related work in distributed data mining} \label{tab:comparison1} \add{\vspace{-0.1in}} \end{table}

1307.0781v1

\begin{table}[t] \centering {\fontsize{8}{6}\selectfont \setlength{\tabcolsep}{.25em} \vspace{-0.2in} \begin{tabular}{|l|c|c|c|c|c|} \hline &\cite{slivkins2009contextual, dudik2011efficient, langford2007epoch, chu2011contextual} & \cite{hliu1, anandkumar, tekin2012sequencing} & \cite{tekin4} & This work \\ \hline Multi-user & no & yes & yes & yes \\ \hline Cooperative & N/A & yes & no & yes \\ \hline Contextual & yes & no & no & yes \\ \hline Data arrival & arbitrary & i.i.d. or Markovian & i.i.d. & i.i.d or arbitrary \\ process& & & & \\ \hline Regret & sublinear & logarithmic & may be linear & sublinear \\ \hline \end{tabular} } \caption{Comparison with related work in multi-armed bandits} \vspace{-0.35in} \label{tab:comparison2} \end{table}

1609.00904v1

\begin{table} \begin{tabular}{|r||c|c|c|c|} \hline Name & Nom. & Int. & Cont. & Note \\ \hline Mad. & 0 & 500 & 0 & hyper-XOR problem \\ Car. & 18 & 0 & 14 & car auction \\ Home. & 295 & 0 & 1 & real estate \\ Mel. & 178 & 61 & 11 & grant applications \\ Credit & 0 & 6 & 4 & credit risks\\ \hline \end{tabular} \caption{Datasets and their characteristics. Nom = nominal. Int = integer. Cont = continuous.} \label{tbl:datasets} \end{table}

2202.10688v2

\begin{table}[] \centering \caption{Datasets} \resizebox{\columnwidth}{!}{% \begin{tabular}{|l|l|l|l|l|} \hline \multicolumn{1}{|c|}{\textbf{Dataset}} & \multicolumn{1}{c|}{\textbf{Nodes}} & \multicolumn{1}{c|}{\textbf{Edges}} & \multicolumn{1}{c|}{\textbf{Features}} & \multicolumn{1}{c|}{\textbf{Tasks}} \\ \hline Citeseer & 3,312 & 4,732 & 3,703 & 6 \\ \hline Cora & 2,708 & 5,429 & 1,433 & 7 \\ \hline OGBN-Arxiv & 169,343 & 1,166,243 & 128 & 40 \\ \hline OGBN-Product & 2,449,029 & 61,859,140 & 100 & 47 \\ \hline OGBN-Proteins & 132,534 & 39,561,252 & 8 & 112 \\ \hline Pubmed & 19,717 & 44,338 & 500 & 3 \\ \hline Reddit & 232,965 & 11,606,919 & 602 & 41 \\ \hline \end{tabular}% } \end{table}

2201.01288v2

\begin{table}[h] \caption{Supported generators in the auto feature engineering module.} \label{tab:generators} \centering \begin{tabular}{l|l} \toprule Name & Description \\ \midrule \texttt{graphlet} & Local graphlet numbers\\ \texttt{eigen} & EigenGNN features. \\ \texttt{pagerank} & PageRank scores. \\ \texttt{PYGLocalDegreeProfile} & Local Degree Profile features \\ \texttt{PYGNormalizeFeatures} & Row-normalize all node features \\ \texttt{PYGOneHotDegree} & One-hot encoding of node degrees. \\ \texttt{onehot} & One-hot encoding of node IDs \\ \bottomrule \end{tabular} \end{table}

1905.09849v2

\begin{table}[h] \small \begin{center} \caption{Test statistics $\hat{m}_j$ and their associated confidence intervals of the first and second-order significant features returned by the SFIT algorithm for $\alpha = 0.05$, $\beta = 10^{-2}$.} \label{table:1} \begin{tabularx}{\columnwidth}{ X|X|X } Significant features or pairs & $\hat{m}_j$ & 95\%-Confidence interval \\ \hline $X_1$ & 1.13 & $[1.06,1.21]$ \\ $X_3$ & 0.224 & $[0.185,0.265]$ \\ $(X_1,X_2)$ & 0.01 & $[0.004,0.016]$ \\ $(X_4,X_5)$ & 0.048 & $[0.035,0.060]$ \\ \hline \end{tabularx} \end{center} \end{table}

1905.09849v2

\begin{table}[h] \small \begin{center} \caption{Predictive performance as measured by AUC and balanced accuracy of the neural network trained on the full set of 75 variables and on the subset of 24 variables selected by SFIT.} \label{table:5} \begin{tabularx}{\columnwidth}{ X|X|X } Metric & Neural network on all variables & Neural network on selected variables\\ \hline AUC & 0.775 & 0.765\\ Balanced accuracy & 0.707 & 0.701 \\ \hline \end{tabularx} \end{center} \end{table}

2112.13790v1

\begin{table}[tb] \caption{Statistics of training and test datasets of two subtasks. The third and fourth columns represent the total of examples and the average number of words by example, respectively. The symbols $\$$ represents the sentiment scores. The last columns represent the amount of examples with the sentiment score smaller and bigger (or equal) than zero, respectively.} \label{tab:dataset} \begin{center} \begin{tabular}{|l|c|c|c|c|c|} \hline \textbf{Type} & \textbf{Split} & \textbf{\#Total} & \textbf{Length} & \textbf{$\$<0$} & \textbf{$\$>=0$} \\ \hline \multirow{2}{*}{StockTwits} & train & 934 & 6.4 & 333 & 601 \\ \cline{2-6} & test & 429 & 6.5 & 141 & 288 \\\hline \multirow{2}{*}{Twitter} & train & 766 & 5.7 & 248 & 518\\ \cline{2-6} & test & 365 & 5.4 & 116 & 249\\\hline \multirow{2}{1.4cm}{News headline} & train & 1142 & 9.6 & 451 & 691\\ \cline{2-6} & test & 491 & 9.5 & 203 & 288 \\ \hline \end{tabular} \end{center} \end{table}

2112.13790v1

\begin{table}[tb] \caption{Cosine similarity scores of the version RoBERTa$+5\times$Transf grouped by positive and negative prediction values.} \begin{center} \label{tab:results:values} % Give a unique label \begin{tabular}{|p{3cm}|c|} \hline \textbf{Prediction values} & \textbf{Cosine} \\ \hline \multicolumn{2}{|l|}{\textit{Microblogs}} \\ \hline positive ($>=0$) & 0.853 \\ negative & 0.815 \\ \hline \multicolumn{2}{|l|}{\textit{News statements \& headlines}} \\\hline positive ($>=0$) & 0.860 \\ negative & 0.840 \\ \hline \end{tabular} \end{center} \end{table}

1802.08251v1

\begin{table}[!h] \caption{Summary of tools inputs and outputs.} \label{table:tools} \scalebox{0.75}{ \begin{tabular}{l|l|l} \hline Tool & Input &Output \\ \hline GEOQuery & GEO database ID & Rdata object and .cond file \\ QCNormalization & Raw .CEL affymetrix files and .conf file & Rdata object and plots \\ Import custom data & Expression data in tabular .txt format &Rdata object and plots \\ Limma analysis & Rdata object from GEOQuery or QCNormalization & Rdata Object and HTML report \\ Microarray data meta-analysis & Rdata objects from Limma analyse & HTML report \\ RNA-seq data meta-analysis & Results text files from galaxy deseq2 tool & HTML report \\ \hline \end{tabular} } \end{table}

2308.10194v1

\begin{table}[h] \caption{Observations per center}\label{table:nobservation} \centering \begin{tabular}{|c | c | c|} \toprule Number of centers & Number of observations in each group & each group total \\ \midrule 3 & 698, 476, 326 & 1500 \\ 5 & 492, 368, 276, 208, 156 & 1500 \\ 10 & 307, 250, 208, 172, 143, 118, 98, 81, 67, 56 & 1500 \\ \bottomrule \end{tabular} \end{table}

2302.11751v1

\begin{table} \caption{Complete inspection on ensemble teams for \textit{EMNIST Balanced} dataset with $m=10, K=6$, \textit{noniid-lds} partition.} \label{tab:completeInspection} \begin{tabular}{c|c|c} \hline Method & Rank & Accuracy (\%) \\ \hline \methodName & \textbf{34/1024} & \textbf{84.34} \\ \hline AS & 114/1024 & 83.39 \\ \hline CV & 241/1024 & 82.29 \\ \hline DS & 348/1024 & 80.86 \\ \hline LD & 608/1024 & 77.77 \\ \hline RS & 669/1024 & 76.54 \\ \hline \end{tabular} \end{table}

2302.11751v1

\begin{table}[htbp] % % \centering % % \caption{Complete Inspection on ensemble teams for CIFAR100 dataset with M=10}% % \begin{tabular}{c|c|c} % \hline % Method & Rank & Accuracy (\%) \\ % \hline % \textbf{DeDes} & \textbf{38/1024} & \textbf{84.22} \\ % \hline % AS & 85/1024 & 83.25 \\ % \hline % CVS & 215/1024 & 82.15 \\ % \hline % DS & 658/1024 & 81.26 \\ % \hline % RS & 798/1024 & 78.85 \\ % \hline % LD & 54/1024 & 83.45 \\ % \hline % \end{tabular}% % \label{tab:completeInspection}% % % \end{table}

2105.07636v2

\begin{table}[h] \caption{DOC (DA/OE) parameters CIFAR-10.} \label{tab_Binary_DEEPSVM} \begin{center} \begin{small} \begin{sc} % \tabcolsep = 0.1cm \begin{tabular}{c|cc} \toprule \specialcell{CIFAR-10 \\ (Class)} & $\lambda = \frac{1}{2C} $ & \specialcell{SGD\\Learning Rate} \\ \midrule Airplane & 1.0 & $5 \times 10^{-3}$ \\ Automobile & 0.01& $5 \times 10^{-4}$ \\ Bird &0.5 & $5 \times 10^{-4}$\\ Cat & 1.0 & $10^{-4}$ \\ Deer & 1.0 & $10^{-4}$ \\ Dog & 0.5 & $10^{-3}$ \\ Frog & 0.5& $5\times 10^{-3}$ \\ Horse & 0.01 & $10^{-3}$ \\ Ship & 1.0 & $10^{-3}$ \\ Truck & 0.01 & $10^{-4}$ \\ \bottomrule \end{tabular} \end{sc} \end{small} \end{center} \vskip -0.1in \end{table}

2105.07636v2

\begin{table}[h] \caption{Optimal Model Parameters for the Tabular data sets.}\label{tab_tabularParamDOC3} \vskip 0.15in \begin{center} \resizebox{\columnwidth}{!}{ \begin{small} % \tabcolsep = 0.1cm \begin{tabular}{c|ccccc} \toprule \specialcell{Class} & $\lambda$ & \specialcell{Learning\\Rate} & $ C_U/C$ & Epoch & Batch Size \\ \midrule Thyroid & $10^{-6}$ & $10^{-3}$ & 5.0 & 500 & 100 \\ Arrhythmia & 0.01 & $10^{-3}$ & 0.001 & 300 & 100 \\ Abalone & 0.1 & $10^{-3}$ & 0.01& 300 & 100 \\ \bottomrule \end{tabular} \end{small}} \end{center} \vskip -0.1in \end{table}

1802.07966v2

\begin{table}[!t] \centering \hspace{-0.6cm} \begin{tabular}{l|p{120pt}|p{123pt}|p{125pt}} \hline &The square is above the rectangle.&The square is below the rectangle. &The square is below the rectangle.\\ $x$&The triangle is to the left of the square.&The triangle is to the right of the square.&The triangle is to the right of the square. \\ &Is the rectangle to the right of the triangle?&Is the rectangle to the right of the triangle?&Is the triangle below the rectangle?\\\hline $y$&Yes& No & Yes\\\hline \end{tabular} \caption{A set of examples taken from the Task $17$ of bAbI question answering dataset.} \label{table:dataset} \end{table}

1107.1445v1

\begin{table}[htbp] \centering \begin{tabular}{|c|c|} \hline {\bf Strategy} & {\bf Sequence} \\ \hline \multicolumn{2}{|c|}{Repeated measurements not allowed} \\ \hline ME Sampling & $10~~1~~9~~2~~3~~8~~5~~7~~4~~6$ \\ \hline IM Sampling & $~1~10~~9~~2~~3~~8~~7~~4~~6~~5$ \\ \hline RND Sampling & $~6~10~~3~~1~~5~~9~~8~~2~~4~~7$ \\ \hline \multicolumn{2}{|c|}{Repeated measurements allowed} \\ \hline ME Sampling & $10~~1~10~~1~~1~10~10~~1~~1~~1$ \\ \hline IM Sampling & $~1~10~~1~10~10~~1~10~10~~1~~1$ \\ \hline \end{tabular} \caption{Results for example $1$, the additive noise case. Design sequences yielded by the three strategies for one particular Monte Carlo run} \label{tab:ex1_add_seq} \end{table}

1107.1445v1

\begin{table}[htbp] \centering \begin{tabular}{|c|c|} \hline {\bf Strategy} & {\bf Sequence} \\ \hline \multicolumn{2}{|c|}{Repeated measurements not allowed} \\ \hline ME Sampling & $~5~~6~~7~~4~~1~~3~~8~~9~~2~10$ \\ \hline IM Sampling & $~1~10~~9~~7~~2~~8~~4~~3~~5~~6$ \\ \hline RND Sampling & $~6~~1~~8~~5~10~~9~~4~~2~~3~~7$ \\ \hline \multicolumn{2}{|c|}{Repeated measurements allowed} \\ \hline ME Sampling & $~5~~6~~6~~5~~5~~5~~6~~5~~6~~5$ \\ \hline IM Sampling & $~1~10~~1~10~10~~1~~1~~1~~1~10$ \\ \hline \end{tabular} \caption{Results for example $1$, the multiplicative noise case. Design sequences yielded by the three strategies for one particular Monte Carlo run} \label{tab:ex1_mult_seq} \end{table}

1711.10420v1

\begin{table} \centering \caption{Direction cosines between axes before and after rotation}\label{tab1} \fontsize{10}{14}\selectfont{ \begin{tabular}{c|c|c|c|c} \hline \multicolumn{2}{c|}{\ } & \multicolumn{3}{|c}{Axes before rotation} \\ \cline{3-5} \multicolumn{2}{c|}{\ } & $X_1$ & \ldots & $X_n$ \\ \hline \hline Axes & $X_1'$ & $cos(X_1',X_1)$ & \ldots & $cos(X_1',X_n)$ \\ \cline{2-5} after & \vdots & \vdots & \vdots & \vdots \\ \cline{2-5} rotation & $X_n'$ & $cos(X_n',X_1)$ & \ldots & $cos(X_n',X_n)$ \\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{Transformation of tensors in matrix notation}\label{tab2} \fontsize{10}{14}\selectfont{ \begin{tabular}{c|c|c|c} \hline Rank & New components & Old components & \multirow{2}{*}{Note} \\ of tensor & expressed by old & expressed by new & \\ \hline \hline 0 & $\varphi '=\varphi$ & $\varphi =\varphi '$ & $\varphi$, $\varphi '$ $-$ scalar \\ \hline {1} & {$v'=R v$} & {$v=R^T v'$} & {$v$, $v'$ $-$ vector} \\ \hline \multirow{2}{*}{2} & \multirow{2}{*}{$C'=R C R^T$} & \multirow{2}{*}{$C=R^T C' R$} & $C$, $C'$ $-$ second rank tensor \\ & & & expressed as a square matrix\\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{The data for analysis (in centimeters)}\label{tab3} \fontsize{10}{14}\selectfont{ \begin{tabular}{c|c|c|c|c} \hline Sepal Length & Sepal Width & Petal Length & Petal Width & Class \\ \hline \hline $5.1$ & $3.5$ & $1.4$ & $0.2$ & Iris-setosa \\ \hline $4.9$ & $3$ & $1.4$ & $0.2$ & $\ldots$ \\ \hline $4.7$ & $3.2$ & $1.3$ & $0.2$ & $\ldots$ \\ \hline $\vdots$ & $\vdots$ & $\vdots$ & $\vdots$ & $\vdots$ \\ \hline $6.4$ & $3.2$ & $4.5$ & $1.5$ & Iris-versicolor \\ \hline $6.9$ & $3.1$ & $4.9$ & $1.5$ & $\ldots$ \\ \hline $5.5$ & $2.3$ & $4$ & $1.3$ & $\ldots$ \\ \hline $\vdots$ & $\vdots$ & $\vdots$ & $\vdots$ & $\vdots$ \\ \hline $6.3$ & $3.3$ & $6$ & $2.5$ & Iris-virginica \\ \hline $5.8$ & $2.7$ & $5.1$ & $1.9$ & $\ldots$ \\ \hline $7.1$ & $3$ & $5.9$ & $2.1$ & $\ldots$ \\ \hline $\vdots$ & $\vdots$ & $\vdots$ & $\vdots$ & $\vdots$ \\ \hline \hline $5.845$ & $3.121$ & $3.770$ & $1.199$ & Average \\ \hline $0.833$ & $0.480$ & $1.773$ & $0.763$ & Standard deviation \\ \hline $0.693$ & $0.230$ & $3.143$ & $0.582$ & Variance \\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{The matrix of correlation coefficient (cosines)}\label{tab4} \fontsize{10}{14}\selectfont{ \begin{tabular}{c|c|c|c|c} \hline & Sepal Length & Sepal Width & Petal Length & Petal Width \\ \hline \hline Sepal Length & $1$ & $-0.063$ & $0.866$ & $0.816$ \\ \hline Sepal Width & $-0.063$ & $1$ & $-0.321$ & $-0.300$ \\ \hline Petal Length & $0.866$ & $-0.321$ & $1$ & $0.959$ \\ \hline Petal Width & $0.816$ & $-0.300$ & $0.959$ & $1$ \\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{The matrix of significance levels}\label{tab5} \fontsize{10}{14}\selectfont{ \begin{tabular}{c|c|c|c|c} \hline & Sepal Length & Sepal Width & Petal Length & Petal Width \\ \hline \hline Sepal Length & $0$ & $0.446$ & $0.000$ & $0.000$ \\ \hline Sepal Width & $0.446$ & $0$ & $0.000$ & $0.000$ \\ \hline Petal Length & $0.000$ & $0.000$ & $0$ & $0.000$ \\ \hline Petal Width & $0.000$ & $0.000$ & $0.000$ & $0$ \\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{The matrix of angles given in degrees}\label{tab6} \fontsize{10}{14}\selectfont{ \begin{tabular}{c|c|c|c|c} \hline & Sepal Length & Sepal Width & Petal Length & Petal Width \\ \hline \hline Sepal Length & $0$ & $93.59$ & $30.05$ & $35.29$ \\ \hline Sepal Width & $93.59$ & $0$ & $108.74$ & $107.47$ \\ \hline Petal Length & $30.05$ & $108.74$ & $0$ & $16.44$ \\ \hline Petal Width & $35.29$ & $107.47$ & $16.44$ & $0$ \\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{The matrix of determination coefficients}\label{tab7} \fontsize{10}{14}\selectfont{ \begin{tabular}{c|c|c|c|c} \hline & Sepal Length & Sepal Width & Petal Length & Petal Width \\ \hline \hline Sepal Length & $100.00\%$ & $0.39\%$ & $74.93\%$ & $66.63\%$ \\ \hline Sepal Width & $0.39\%$ & $100.00\%$ & $10.32\%$ & $9.01\%$ \\ \hline Petal Length & $74.93\%$ & $10.32\%$ & $100.00\%$ & $91.99\%$ \\ \hline Petal Width & $66.63\%$ & $9.01\%$ & $91.99\%$ & $100.00\%$ \\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{Eigenvectors in columns}\label{tab8} \fontsize{10}{14}\selectfont{ \begin{tabular}{c|c|c|c} \hline $0.534$ & $0.317$ & $0.757$ & $0.203$ \\ \hline $-0.213$ & $0.948$ & $-0.229$ & $-0.066$ \\ \hline $0.584$ & $0.026$ & $-0.212$ & $-0.783$ \\ \hline $0.573$ & $0.030$ & $-0.574$ & $0.584$ \\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{Several objects in the principal components space}\label{tab9} \fontsize{10}{14}\selectfont{ \begin{tabular}{c|c|c|c|c} \hline No. & $p_{c1}$ & $p_{c2}$ & $p_{c3}$ & $p_{c4}$ \\ \hline \hline $1$ & $-2.184$ & $0.393$ & $-0.176$ & $0.048$ \\ \hline $2$ & $-2.091$ & $-0.674$ & $-0.233$ & $0.068$ \\ \hline $3$ & $-2.341$ & $-0.356$ & $0.033$ & $0.036$ \\ \hline $\vdots$ & $\vdots$ & $\vdots$ & $\vdots$ & $\vdots$ \\ \hline \hline Average & $0.000$ & $0.000$ & $0.000$ & $0.000$ \\ \hline Standard deviation & $1.694$ & $0.984$ & $0.398$ & $0.181$ \\ \hline Variance & $2.868$ & $0.967$ & $0.159$ & $0.033$ \\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{The percentage of variance explained by the successive principal components (in brief: PC)}\label{tab10} \fontsize{10}{14}\selectfont{ \begin{tabular}{c|c|c|c|c} \hline \multirow{2}{*}{No.} & \multirow{2}{*}{Eigenvalue} & Cumulative & Percentage of variance & Cumulative \\ & & eigenvalues & explained by each PC & percentage of variance \\ \hline \hline $1$ & $2.849$ & $2.849$ & $71.22\%$ & $71.22\%$ \\ \hline $2$ & $0.961$ & $3.810$ & $24.02\%$ & $95.24\%$ \\ \hline $3$ & $0.158$ & $3.967$ & $3.94\%$ & $99.19\%$ \\ \hline $4$ & $0.033$ & $4.000$ & $0.81\%$ & $100.00\%$ \\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{Correlation coefficients between primary variables and principal components}\label{tab11} \fontsize{10}{14}\selectfont{ \begin{tabular}{c||c|c|c|c} \hline & Sepal Length & Sepal Width & Petal Length & Petal Width \\ \hline \hline $p_{c1}$ & $0.901$ & $-0.359$ & $0.986$ & $0.968$ \\ \hline $p_{c2}$ & $0.311$ & $0.929$ & $0.025$ & $0.030$ \\ \hline $p_{c3}$ & $-0.301$ & $0.091$ & $0.084$ & $0.228$ \\ \hline $p_{c4}$ & $0.037$ & $-0.012$ & $-0.141$ & $0.105$ \\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{Determination coefficients between primary variables and principal components}\label{tab12} \fontsize{10}{14}\selectfont{ \begin{tabular}{c||c|c|c|c||c} \hline &Sepal Length&Sepal Width&Petal Length&Petal Width & $\Sigma $ \\ \hline \hline $p_{c1} $ & $0.812$ & $0.129$ & $0.972$ & $0.936$ & $2.849$ \\ \hline $p_{c2} $ & $0.097$ & $0.863$ & $0.001$ & $0.001$ & $0.961$ \\ \hline $p_{c3} $ & $0.090$ & $0.008$ & $0.007$ & $0.052$ & $0.158 $\\ \hline $p_{c4} $ & $0.001$ & $0.000$ & $0.020$ & $0.011$ & $0.033 $ \\ \hline \hline $\Sigma$ & $1.000$ & $1.000$ & $1.000$ & $1.000$ & $4.000$ \\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{Level of reconstruction of primary variables}\label{tab13} \fontsize{10}{14}\selectfont{ \begin{tabular}{c||c|c|c|c||c} \hline &Sepal Length&Sepal Width&Petal Length&Petal Width & Average in row \\ \hline \hline $p_{c1} $ & $81.17\%$ & $12.88\%$ & $97.23\%$ & $93.61\%$ & $71.22\%$ \\ \hline $p_{c2} $ & $9.65\%$ & $86.28\%$ & $0.06\%$ & $0.09\%$ & $24.02\%$ \\ \hline \hline $\Sigma$ & $90.82\%$ & $99.16\%$ & $97.29\%$ & $93.70\%$ & $95.24\%$ \\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{The similarity of the primary variables and selected principal components measured by the coefficient of determination }\label{tab14} \fontsize{10}{14}\selectfont{ \begin{tabular}{c||c|c|c|c} \hline &Sepal Length&Sepal Width&Petal Length&Petal Width \\ \hline \hline Similarity to $p_{c1}$ & $0.812$ & $0.129$ & $0.972$ & $0.936$\\ \hline Similarity to $p_{c2}$ & $0.097$ & $0.863$ & $0.001$ & $0.001$\\ \hline \end{tabular}} \end{table}

1711.10420v1

\begin{table} \centering \caption{Operations on tensors}\label{tab15} \fontsize{10}{14}\selectfont{ \begin{tabular}{c|c|c} \hline Transition &Back &\multirow{3}{*}{Note} \\ to the new& to the old&\\ coordinate system&coordinate system & \\ \hline\hline \multirow{3}{*}{$A'=RA$} & \multirow{3}{*}{$A=R^T A'$} & Standardized primary variables \\ & & are represented as vectors \\ & & (columns in matrices $A$ and $A'$)\\ \hline \multirow{3}{*}{$P'=RP$} & \multirow{3}{*}{$P=R^T P'$} & Principal components \\ & & are represented as vectors \\ & & (columns in matrices $P$ and $P'$)\\ \hline \multirow{2}{*}{$C'=RCR^T $} &\multirow{2}{*}{$C=R^T C' R$} & The correlation coefficient matrix \\ & & ($C$, $C'$) is a tensor of rank two \\ \hline \end{tabular}} \end{table}

2205.01168v1

\begin{table} \scriptsize \centering \caption{ Performance summary and data statistics of the largest end-to-end runs for concatenating the ND and FD files. With the tuned chunk size, the compression ratio is significantly improved over the sum of the input files, as shown in Table \ref{table:statistics}. The overall metadata size is also reduced. } \begin{tabular}{|c||c|c|} \hline Data & 165 ND files & 6,400 FD files \\ \hline \hline Number of processes (nodes) & 165 (42) & 1600 (400) \\ \hline Timing & 279.4 sec & 611.8 sec \\ \hline Data size before compression & 1 {\bf TB} & 16.9 {\bf TB} \\ \hline Output file size & 15.1 GB & 77.9 GB \\ \hline Metadata size & 57.7 MB & 93.8 MB \\ \hline Raw data size & 15.0 GB & 77.4 GB \\ \hline \end{tabular} \label{table:summary1} \end{table}

2205.01168v1

\begin{table} \scriptsize \centering \caption{NOvA data table organization with one entry per slice.} \begin{tabular}{|r|r|r|r|r|r|} \hline Run & Subrun & Event & Sub- & distallpngtop & ... 35 more \\ & & & event & & ... \\ \hline \hline 433 & 61 & 6124 & 35 & nan & \\ \hline 433 & 61 & 6124 & 36 & -0.7401 & \\ \hline 433 & 61 & 6124 & 37 & nan & \\ \hline 433 & 61 & 6125 & 1 & nan & \\ \hline 433 & 61 & 6125 & 2 & 423.633 & \\ \hline 433 & 61 & 6125 & 3 & -2.8498 & \\ \hline \hline \end{tabular} \label{table:slice} \end{table}

2205.01168v1

\begin{table} \centering \scriptsize \caption{NOvA data table organization with one entry per vertex.} \begin{tabular}{|r|r|r|r|r|r|r|} \hline Run & Subrun & Event & Sub- & vtxid & npng3d & ... 6 more \\ & & & event & & & ... \\ \hline \hline 433 & 61 & 6124 & 35 & 0 & 0 & \\ \hline 433 & 61 & 6124 & 36 & 0 & 1 & \\ \hline 433 & 61 & 6124 & 36 & 1 & 1 & \\ \hline 433 & 61 & 6124 & 36 & 2 & 5 & \\ \hline 433 & 61 & 6125 & 1 & 0 & 1 & \\ \hline 433 & 61 & 6125 & 3 & 0 & 0 & \\ \hline \hline \end{tabular} \label{table:vertex} \end{table}

2205.01168v1

\begin{table} \scriptsize \centering \caption{Statistics of NOvA ND and FD data files.} \begin{tabular}{|r||r|r|} \hline & ND data files & FD data files \\ \hline \hline \# of files & 165 & 6,400 \\ \hline \# of groups per file & 999 & 701 \\ \hline \# of 1D datasets per file & 15,965 & 12,925 \\ \hline \# of 2D datasets per file & 8 & 6 \\ \hline \# of empty datasets & 13,396 & 9,374 \\ \hline \hline Compression & GZIP-level 6 & GZIP-level 6 \\ \hline Chunk size & 128-element based & 128-element based \\ \hline 1D dsets before compr. & 97.9 GB & 413.3 GB \\ \hline 1D dsets after compr. & 21.4 GB & 69.8 GB \\ \hline 2D dsets before compr. & 903.2 GB & 16.6 \textbf{\textit{TB}} \\ \hline 2D dsets after compr. & 2.1 GB & 32.1 GB \\ \hline Overall before compr. & 1001.1 GB & 17.0 \textbf{\textit{TB}} \\ \hline Overall after compr. & 23.5 GB & 101.9 GB \\ \hline \hline Total file size & 35.2 GB & 212.5 GB \\ \hline Metadata size & 11.7 GB & 104.1 GB\\ \hline Raw data size & 23.5 GB & 108.1 GB \\ \hline \end{tabular} \label{table:statistics} \end{table}

2205.01168v1

\begin{table} \scriptsize \centering \caption{ Maximum number of rounds of reads and writes among all the datasets for different I/O buffer sizes. The total number of MPI processes and input ND files is fixed at 128, while the number of processes per node varies between 4 and 32. The I/O buffer size allocated in each process is 8 GB when running 4 processes per node, and it proportionally decreases as the number of processes run on each node increases. } \begin{tabular}{|r|r|r|r|r|r|r|r|} \hline \multicolumn{2}{|r|}{Number of processes} & \multirow{2}{*}{4} & \multirow{2}{*}{8} & \multirow{2}{*}{16} & \multirow{2}{*}{32} & \multirow{2}{*}{64} & \multirow{2}{*}{128} \\ \cline{1-2} procs per node & I/O buffer & & & & & & \\ \hline \hline 4 & 8 GB & 24 & 12 & 6 & 4 & 2 & 1 \\ \hline 8 & 4 GB & - & 24 & 12 & 7 & 4 & 2 \\ \hline 16 & 2 GB & - & - & 24 & 13 & 7 & 4 \\ \hline 32 & 1 GB & - & - & - & 25 & 13 & 7 \\ \hline \end{tabular} \label{table:nproc} \end{table}