_id
sequence
label
stringlengths
5
77
text
stringlengths
78
3.43k
[ 0 ]
cs.CV
Fashion Meets Computer Vision A Survey. Fashion is the way we present ourselves to the world and has become one of the world's largest industries. Fashion, mainly conveyed by vision, has thus attracted much attention from computer vision researchers in recent years. Given the rapid development, this paper provides a comprehensive survey of more than 200 major fashion related works covering four main aspects for enabling intelligent fashion (1) Fashion detection includes landmark detection, fashion parsing, and item retrieval, (2) Fashion analysis contains attribute recognition, style learning, and popularity prediction, (3) Fashion synthesis involves style transfer, pose transformation, and physical simulation, and (4) Fashion recommendation comprises fashion compatibility, outfit matching, and hairstyle suggestion. For each task, the benchmark datasets and the evaluation protocols are summarized. Furthermore, we highlight promising directions for future research.
[ 1, 3 ]
cs.CL cs.LG
Semi supervised Relation Extraction via Incremental Meta Self Training. To alleviate human efforts from obtaining large scale annotations, Semi Supervised Relation Extraction methods aim to leverage unlabeled data in addition to learning from limited samples. Existing self training methods suffer from the gradual drift problem, where noisy pseudo labels on unlabeled data are incorporated during training. To alleviate the noise in pseudo labels, we propose a method called MetaSRE, where a Relation Label Generation Network generates quality assessment on pseudo labels by (meta) learning from the successful and failed attempts on Relation Classification Network as an additional meta objective. To reduce the influence of noisy pseudo labels, MetaSRE adopts a pseudo label selection and exploitation scheme which assesses pseudo label quality on unlabeled samples and only exploits high quality pseudo labels in a self training fashion to incrementally augment labeled samples for both robustness and accuracy. Experimental results on two public datasets demonstrate the effectiveness of the proposed approach.
[ 25 ]
math.PR
Long Time Influence of Small Perturbations and Motion on the Simplex of Invariant Probability Measures. A general approach to a broad class of asymptotic problems related to long time influence of small perturbations, of both deterministic and stochastic type, is presented in the paper. The main characteristic of this influence is a limiting motion on the simplex of invariant probability measures of the non perturbed system in an appropriate time scale. The main tools we use in the paper are limit theorems for large deviations, modified averaging principle, and diffusion approximation.
[ 19 ]
cs.CR
A Look at the Dark Side of Hardware Reverse Engineering A Case Study. A massive threat to the modern and complex IC production chain is the use of untrusted off shore foundries which are able to infringe valuable hardware design IP or to inject hardware Trojans causing severe loss of safety and security. Similarly, market dominating SRAM based FPGAs are vulnerable to both attacks since the crucial gate level netlist can be retrieved even in field for the majority of deployed device series. In order to perform IP infringement or Trojan injection, reverse engineering (parts of) the hardware design is necessary to understand its internal workings. Even though IP protection and obfuscation techniques exist to hinder both attacks, the security of most techniques is doubtful since realistic capabilities of reverse engineering are often neglected. The contribution of our work is twofold first, we carefully review an IP watermarking scheme tailored to FPGAs and improve its security by using opaque predicates. In addition, we show novel reverse engineering strategies on proposed opaque predicate implementations that again enables to automatically detect and alter watermarks. Second, we demonstrate automatic injection of hardware Trojans specifically tailored for third party cryptographic IP gate level netlists. More precisely, we extend our understanding of adversary's capabilities by presenting how block and stream cipher implementations can be surreptitiously weakened.
[ 65 ]
cs.NI
Traffic Analysis for Storage Finding in Video on Demand System. The literature survey typically predicated sharp growth for IP based video traffic i.e., 30 or more annually. For the Internet TV in mobile networks, video traffic growth rate is expected to rise 80 or more. These high growth rates of video traffic will account for a large portion of the bandwidth. The performance of video on demand system during real time data streaming greatly depends on the session oriented data storage finding in the mass scale distributed storage architecture. At the storage end, data is broken up into manageable chunks of data packets, which could be smoothly, deliver over the Internet. The objective of this study was to present the necessity of traffic control and traffic analysis methodology in the video on demand system to minimize the hop count for finding exact media storage to retrieve video chunk data.
[ 0, 1 ]
cs.CV cs.LG
Multi Precision Policy Enforced Training (MuPPET) A precision switching strategy for quantised fixed point training of CNNs. Large scale convolutional neural networks (CNNs) suffer from very long training times, spanning from hours to weeks, limiting the productivity and experimentation of deep learning practitioners. As networks grow in size and complexity, training time can be reduced through low precision data representations and computations. However, in doing so the final accuracy suffers due to the problem of vanishing gradients. Existing state of the art methods combat this issue by means of a mixed precision approach utilising two different precision levels, FP32 (32 bit floating point) and FP16 FP8 (16 8 bit floating point), leveraging the hardware support of recent GPU architectures for FP16 operations to obtain performance gains. This work pushes the boundary of quantised training by employing a multilevel optimisation approach that utilises multiple precisions including low precision fixed point representations. The novel training strategy, MuPPET, combines the use of multiple number representation regimes together with a precision switching mechanism that decides at run time the transition point between precision regimes. Overall, the proposed strategy tailors the training process to the hardware level capabilities of the target hardware architecture and yields improvements in training time and energy efficiency compared to state of the art approaches. Applying MuPPET on the training of AlexNet, ResNet18 and GoogLeNet on ImageNet (ILSVRC12) and targeting an NVIDIA Turing GPU, MuPPET achieves the same accuracy as standard full precision training with training time speedup of up to 1.84 times and an average speedup of 1.58 times across the networks.
[ 9 ]
math.AP
Dual variational methods for a nonlinear Helmholtz equation with sign changing nonlinearity. We prove new existence results for a Nonlinear Helmholtz equation with sign changing nonlinearity of the form Delta u k 2 u Q(x) u p 2 u, quad u in W 2,p ( mathbb R N ) with k 0, N geq 3 , p in left left. frac 2(N 1) N 1 , frac 2N N 2 right) right. and Q in L infty ( mathbb R N ) . Due to the sign changes of Q , our solutions have infinite Morse Index in the corresponding dual variational formulation.
[ 42 ]
math.DG
On Cohomogeneity One Linear Actions On Pseudo euclidean Space mathbb R p,q . The aim of this paper is to study cohomogeneity one isometric linear actions on the p q dimensional pseudo Euclidean space mathbb R p,q . It is proved that the natural isometric action of the nilpotent factor of an Iwasawa decomposition of SO(p,q) is not of cohomogeneity one. The orbits of cohomogeneity one actions of some subgroups of a maximal parabolic subgroup of the isometry group of mathbb R p,q are determined and it is proved that there exist cohomogeneity one isometric actions on mathbb R p,q which are orbit equivalent on the complement of a p dimensional degenerate subspace mathbb W p of mathbb R p,q and not orbit equivalent on mathbb W p .
[ 6 ]
physics.flu-dyn
Non ideal gas dynamics under confinement rarefaction effect, dense effect and molecular interaction. The effects of volume exclusion and long range intermolecular attraction are investigated by the simplified kinetic model for surface confined inhomogeneous fluids. Gas dynamics of the ideal gas, the hard sphere fluid and the real gas are simulated by the Boltzmann equation, the Enskog equation and the simple kinetic equation, respectively. Only the Knudsen minimum appears for the ideal gas, while both the Knudsen minimum and the Knudsen maximum occur for the hard sphere fluid and the real gas under certain confinements, beyond which the maximum and minimum may disappear. The Boltzmann equation and the Enskog equation overestimates and underestimates the mass flow rate of the real gas dynamics under confinement, respectively, where the volume exclusion and the long range intermolecular attractive potential among molecules are not ignorable. With the increase of the channel width, gas dynamics of the hard sphere fluid and the real gas tends to the Boltzmann prediction gradually. The density inhomogeneity, which hinders the flow under confinement, is more obvious when the solid fraction is larger. The anomalous slip occurs for real gas under constant confinement. The flow at a smaller Knudsen number (larger solid fraction or channel width) contributes more practical amount of mass transfer, although the rarefaction effects is more prominent at larger Knudsen numbers. The temperature has no effect on density and velocity profiles of the ideal gas and the hard sphere fluid, but the energy parameter among the real gas molecules decreases with the increasing temperature and the real gas dynamics tends to the hard sphere ones consequently.
[ 24, 47 ]
math.MG math.AT
Coarse Homotopy on metric Spaces and their Corona. This paper discusses properties of the Higson corona by means of a quotient on coarse ultrafilters on a proper metric space. We use this description to show that the corona functor is faithful. This study provides a K "unneth formula for twisted coarse cohomology. We obtain the Gromov boundary of a hyperbolic proper geodesic metric space as a quotient of its Higson corona.
[ 24, 21 ]
math.CO math.MG
Quasi Euclidean classification of Alcoved Convex Polyhedra. We give the quasi Euclidean classification of the maximal (with respect to the f vector) alcoved polyhedra. The f vector of these maximal convex bodies is (20,30,12) , so they are simple dodecahedra. We find eight quasi Euclidean classes. This classification, which preserves angles, is finer than the known combinatorial classification (found in 2012 by Jim ' e nez and de la Puente), which has only six classes. Each alcoved polyhedron mathcal P is represented by a unique visualized idempotent matrix A . Some 2 minors of A are invariants of mathcal P they are the tropical edge lengths of mathcal P .
[ 3 ]
cs.CL
The Style Content Duality of Attractiveness Learning to Write Eye Catching Headlines via Disentanglement. Eye catching headlines function as the first device to trigger more clicks, bringing reciprocal effect between producers and viewers. Producers can obtain more traffic and profits, and readers can have access to outstanding articles. When generating attractive headlines, it is important to not only capture the attractive content but also follow an eye catching written style. In this paper, we propose a Disentanglement based Attractive Headline Generator (DAHG) that generates headline which captures the attractive content following the attractive style. Concretely, we first devise a disentanglement module to divide the style and content of an attractive prototype headline into latent spaces, with two auxiliary constraints to ensure the two spaces are indeed disentangled. The latent content information is then used to further polish the document representation and help capture the salient part. Finally, the generator takes the polished document as input to generate headline under the guidance of the attractive style. Extensive experiments on the public Kuaibao dataset show that DAHG achieves state of the art performance. Human evaluation also demonstrates that DAHG triggers 22 more clicks than existing models.
[ 25 ]
math.PR
Formulae for the derivatives of heat semigroups. We use a basic martingale method to show a differentiation formula for the derivatives d(P tf)(x 0)(v 0) 1 over t E f(x t) int 0 t langle Y(x s)(v s),dB t rangle R m . These are proved first on R n , then on manifolds. Afterwards for solutions of heat equations on differential forms, and a second order formula.
[ 24, 12 ]
math.CA math.MG
A d dimensional Analyst's Travelling Salesman Theorem for general sets in mathbb R n . In his 1990 paper, Jones proved the following given E subseteq mathbb R 2 , there exists a curve Gamma such that E subseteq Gamma and mathscr H 1( Gamma) sim text diam , E sum Q beta E (3Q) 2 ell(Q). Here, beta E(Q) measures how far E deviates from a straight line inside Q . This was extended by Okikiolu to subsets of mathbb R n and by Schul to subsets of a Hilbert space. In 2018, Azzam and Schul introduced a variant of the Jones beta number. With this, they, and separately Villa, proved similar results for lower regular subsets of mathbb R n. In particular, Villa proved that, given E subseteq mathbb R n which is lower content regular, there exists a nice' d dimensional surface F such that E subseteq F and begin align mathscr H d(F) sim text diam ( E) d sum Q beta E (3Q) 2 ell(Q) d. end align In this context, a set F is nice' if it satisfies a certain topological non degeneracy condition, first introduced in a 2004 paper of David. In this paper we drop the lower regularity condition and prove an analogous result for general d dimensional subsets of mathbb R n. To do this, we introduce a new d dimensional variant of the Jones beta number that is defined for any set in mathbb R n.
[ 33 ]
physics.ins-det
Development of a 3 D energy momentum analyzer for meV scale energy electrons. In this article, we report on the development of a time of flight based electron energy analyzer capable of measuring the 3 D momentum and energy distributions of very low energy (meV scale) photoemitted electrons. This analyzer is capable of measuring energy and 3 D momentum distributions of electrons with energies down to 1 meV with a sub meV energy resolution. This analyzer is an ideal tool for studying photoemission processes very close to the photoemission threshold and also for studying the physics of photoemission based electron sources.
[ 20 ]
math.NA cs.NA
Implicit algorithms for eigenvector nonlinearities. We study and derive algorithms for nonlinear eigenvalue problems, where the system matrix depends on the eigenvector, or several eigenvectors (or their corresponding invariant subspace). The algorithms are derived from an implicit viewpoint. More precisely, we change the Newton update equation in a way that the next iterate does not only appear linearly in the update equation. Although, the modifications of the update equation make the methods implicit we show how corresponding iterates can be computed explicitly. Therefore we can carry out steps of the implicit method using explicit procedures. In several cases, these procedures involve a solution of standard eigenvalue problems. We propose two modifications, one of the modifications leads directly to a well established method (the self consistent field iteration) whereas the other method is to our knowledge new and has several attractive properties. Convergence theory is provided along with several simulations which illustrate the properties of the algorithms.
[ 0 ]
cs.CV
Steering Self Supervised Feature Learning Beyond Local Pixel Statistics. We introduce a novel principle for self supervised feature learning based on the discrimination of specific transformations of an image. We argue that the generalization capability of learned features depends on what image neighborhood size is sufficient to discriminate different image transformations The larger the required neighborhood size and the more global the image statistics that the feature can describe. An accurate description of global image statistics allows to better represent the shape and configuration of objects and their context, which ultimately generalizes better to new tasks such as object classification and detection. This suggests a criterion to choose and design image transformations. Based on this criterion, we introduce a novel image transformation that we call limited context inpainting (LCI). This transformation inpaints an image patch conditioned only on a small rectangular pixel boundary (the limited context). Because of the limited boundary information, the inpainter can learn to match local pixel statistics, but is unlikely to match the global statistics of the image. We claim that the same principle can be used to justify the performance of transformations such as image rotations and warping. Indeed, we demonstrate experimentally that learning to discriminate transformations such as LCI, image warping and rotations, yields features with state of the art generalization capabilities on several datasets such as Pascal VOC, STL 10, CelebA, and ImageNet. Remarkably, our trained features achieve a performance on Places on par with features trained through supervised learning with ImageNet labels.
[ 66 ]
cs.CE
Simultaneous topology and fastener layout optimization of assemblies considering joint failure. This paper provides a method for the simultaneous topology optimization of parts and their corresponding joint locations in an assembly. Therein, the joint locations are not discrete and predefined, but continuously movable. The underlying coupling equations allow for connecting dissimilar meshes and avoid the need for remeshing when joint locations change. The presented method models the force transfer at a joint location not only by using single spring elements but accounts for the size and type of the joints. When considering riveted or bolted joints, the local part geometry at the joint location consists of holes that are surrounded by material. For spot welds, the joint locations are filled with material and may be smaller than for bolts. The presented method incorporates these material and clearance zones into the simultaneously running topology optimization of the parts. Furthermore, failure of joints may be taken into account at the optimization stage, yielding assemblies connected in a fail safe manner.
[ 1, 7 ]
physics.comp-ph cs.LG
Fast Modeling and Understanding Fluid Dynamics Systems with Encoder Decoder Networks. Is a deep learning model capable of understanding systems governed by certain first principle laws by only observing the system's output? Can deep learning learn the underlying physics and honor the physics when making predictions? The answers are both positive. In an effort to simulate two dimensional subsurface fluid dynamics in porous media, we found that an accurate deep learning based proxy model can be taught efficiently by a computationally expensive finite volume based simulator. We pose the problem as an image to image regression, running the simulator with different input parameters to furnish a synthetic training dataset upon which we fit the deep learning models. Since the data is spatiotemporal, we compare the performance of two alternative treatments of time a convolutional LSTM versus an autoencoder network that treats time as a direct input. Adversarial methods are adopted to address the sharp spatial gradient in the fluid dynamic problems. Compared to traditional simulation, the proposed deep learning approach enables much faster forward computation, which allows us to explore more scenarios with a much larger parameter space given the same time. It is shown that the improved forward computation efficiency is particularly valuable in solving inversion problems, where the physics model has unknown parameters to be determined by history matching. By computing the pixel level attention of the trained model, we quantify the sensitivity of the deep learning model to key physical parameters and hence demonstrate that the inversion problems can be solved with great acceleration. We assess the efficacy of the machine learning surrogate in terms of its training speed and accuracy. The network can be trained within minutes using limited training data and achieve accuracy that scales desirably with the amount of training data supplied.
[ 18, 21 ]
math.GT math.CO
Smooth multisections of odd dimensional tori and other manifolds. After Gay and Kirby extended the classical notion of Heegaard splittings of 3 manifolds by introducing trisections of smooth 4 manifolds, Rubinstein and Tillmann defined multisections of PL n manifolds. We consider smooth multisections, which are decompositions into k lfloor n 2 rfloor 1 n dimensional 1 handlebodies with nice intersection properties. We prove that a manifold X of even dimension n geq 6 admits no smooth multisection if H i(X) neq 0 for any odd i neq 1,n 1 . By contrast, every manifold admits a PL multisection, and every manifold of dimension n leq 5 admits a smooth multisection. What about odd dimensions n geq 7 ? We construct, for each odd dimensional torus T n , a smooth multisection which is efficient in the sense that each 1 handlebody has genus n , which we prove is optimal each multisection is symmetric with respect to both the permutation action of S n on the indices and the mathbb Z k translation action along the main diagonal. We also construct such a trisection of T 4 , lift all symmetric multisections of tori to certain cubulated manifolds, and obtain combinatorial identities as corollaries.
[ 42, 18, 76, 13 ]
math.DS math.DG math.GT math.SP
Dynamical torsion for contact Anosov flows. We introduce a new object, the dynamical torsion, which extends the potentially ill defined value at 0 of the Ruelle zeta function of a contact Anosov flow twisted by an acyclic representation of the fundamental group. We show important properties of the dynamical torsion it is invariant under deformations among contact Anosov flows, it is holomorphic in the representation and it has the same logarithmic derivative as some refined combinatorial torsion of Turaev. This shows that the ratio between this torsion and the Turaev torsion is locally constant on the space of acyclic representations. In particular, for contact Anosov flows path connected to the geodesic flow of some hyperbolic manifold among contact Anosov flows, we relate the leading term of the Laurent expansion of zeta at the origin, the Reidemeister torsion and the torsions of the finite dimensional complexes of the generalized resonant states of both flows for the resonance 0 . This extends previous work of cite dang2018fried on the Fried conjecture near geodesic flows of hyperbolic 3 manifolds, to hyperbolic manifolds of any odd dimension.
[ 84 ]
physics.acc-ph
Dynamics of Intense Particle Beam in Axial Symmetric Magnetic Field. Axial symmetric magnetic field is often used in focusing of particle beams. Most existing ion Low Energy Beam Transport lines are based on solenoid focusing. Modern accelerator projects utilize superconducting solenoids in combination with superconducting accelerating cavities for acceleration of high intensity particle beams. Present paper discusses conditions for matched beam in axial symmetric magnetic field. Analysis allows us to minimize power consumption of solenoids and beam emittance growth due to nonlinear space charge, lens aberrations, and maximize acceptance of the channel. Expressions for maximum beam current in focusing structure, beam emittance growth due to spherical aberrations and non linear space charge forces are derived.
[ 42 ]
math.DG
Conformal eta Ricci Soliton on Sasakian manifold. In this paper we study Conformal eta Ricci soliton on Sasakian manifolds. Here, we discuss some curvature properties on Sasakian manifold admitting Conformal eta Ricci soliton. We obtain some significant results on Conformal eta Ricci soliton in Sasakian manifolds satisfying R( xi ,X).S 0, S( xi ,X).R 0, overline P( xi ,X).S 0, where overline P is Pseudo projective curvature tensor.The conditions for Conformal eta Ricci soliton on Phi conharmonically flat and Phi projectively flat Sasakian manifolds have been obtained in this article. Lastly we give an example of 5 dimensional Sasakian manifolds satisfying Conformal eta Ricci soliton.
[ 8 ]
physics.optics
Spectral and temporal modulations of femtosecond SPP wave packets induced by resonant transmission reflection interactions with metal insulator metal nanocavities. To study the dynamical optical interactions of nano scaled metal insulator metal (MIM) structures in temporal frequency domain, femtosecond surface plasmon polariton (SPP) wave packets propagate over a surface with a MIM structure. The resonance nature of the SPP cavity interaction is reflected as strong modulations in the spectra of transmitted and reflected SPP wavepackets, which show peaks and valleys, respectively, corresponding to the MIM cavity's eigenmode. These features indicate that the MIM structure acts as a Fabry Perot etalon type spectrum filter. With appropriate tuning of the resonance frequency of the cavity, one can extract a wave packet with a narrower time duration and temporally shifted intensity peak.
[ 9, 42 ]
math.AP math.DG
Rigidity and Flexibility of Isometric Extensions. In this paper we consider the rigidity and flexibility of C 1, theta isometric extensions and we show that the H "older exponent theta 0 frac12 is critical in the following sense if u in C 1, theta is an isometric extension of a smooth isometric embedding of a codimension one submanifold Sigma and theta frac12 , then the tangential connection agrees with the Levi Civita connection along Sigma . On the other hand, for any theta frac12 we can construct C 1, theta isometric extensions via convex integration which violate such property. As a byproduct we get moreover an existence theorem for C 1, theta isometric embeddings, theta frac12 , of compact Riemannian manifolds with C 1 metrics and sharper amount of codimension.
[ 5 ]
physics.plasm-ph
High charge plasma electron bunches from two collinear laser pulses. An optical injection scheme into the laser wakefield accelerator by preceding injection pulse is investigated by means of 3D numerical particle in cell simulations. Quasimonoenergetic hundred pC electron bunches as short as 6 fs can be generated. Optimal beam separation distance is found at the intersection point of the injection beam bubble with the collection volume for transverse injection into the accelerator beam bubble. It approximately corresponds to the plasma wavelength. The main advantage of this scheme is the localized injection of high charge. This injection mechanism can be useful for applications such as ultrashort and relatively intense X ray radiation sources such as a betatron radiation or Thomson backscattering, time resolved electron diffraction or for seeding of further acceleration stages.
[ 35, 69 ]
math.RA math.QA
Fixed rings of quantum generalized Weyl algebras. Generalized Weyl Algebras (GWAs) appear in diverse areas of mathematics including mathematical physics, noncommutative algebra, and representation theory. We study the invariants of quantum GWAs under finite automorphisms. We extend a theorem of Jordan and Wells and apply it to determine the fixed ring of quantum GWAs under diagonal automorphisms. We further study properties of the fixed rings, including global dimension, rigidity, and simplicity.
[ 0 ]
cs.CV
TunaGAN Interpretable GAN for Smart Editing. In this paper, we introduce a tunable generative adversary network (TunaGAN) that uses an auxiliary network on top of existing generator networks (Style GAN) to modify high resolution face images according to user's high level instructions, with good qualitative and quantitative performance. To optimize for feature disentanglement, we also investigate two different latent space that could be traversed for modification. The problem of mode collapse is characterized in detail for model robustness. This work could be easily extended to content aware image editor based on other GANs and provide insight on mode collapse problems in more general settings.
[ 29, 77, 13 ]
math.DS math.GR math.LO
Minimal model universal flows for locally compact Polish groups. Let G be a locally compact Polish group. A metrizable G flow Y is called model universal if by considering the various invariant probability measures on Y , we can recover every free action of G on a standard Lebesgue space up to isomorphism. Weiss has shown that for countable G , there exists a minimal, model universal flow. In this paper, we extend this result to all locally compact Polish groups.
[ 64, 31 ]
q-bio.PE physics.soc-ph
Shut and re open the role of schools in the spread of COVID 19 in Europe. We investigate the effect of school closure and subsequent reopening on the transmission of COVID 19, by considering Denmark, Norway, Sweden, and German states as case studies. By comparing the growth rates in daily hospitalisations or confirmed cases under different interventions, we provide evidence that the effect of school closure is visible as a reduction in the growth rate approximately 9 days after implementation. Limited school attendance, such as older students sitting exams or the partial return of younger year groups, does not appear to significantly affect community transmission. A large scale reopening of schools while controlling or suppressing the epidemic appears feasible in countries such as Denmark or Norway, where community transmission is generally low. However, school reopening can contribute to significant increases in the growth rate in countries like Germany, where community transmission is relatively high. Our findings underscore the need for a cautious evaluation of reopening strategies that ensure low classroom occupancy and a solid infrastructure to quickly identify and isolate new infections.
[ 25 ]
math.PR
Passage time of the frog model has a sublinear variance. In this paper, we show that the first passage time in the frog model on Z d with d geq 2 has a sublinear variance. This implies that the central limit theorem does not holds at least with the standard diffusive scaling. The proof is based on the method introduced in cite BRo, DHS combining with a control of the maximal weight of paths in locally dependent site percolation. We also apply this method to get the linearity of the lengths of optimal paths..
[ 94 ]
physics.ed-ph
A Conceptual Blend Analysis of Physics Quantitative Literacy Reasoning Inventory Items. Mathematical reasoning flexibility across physics contexts is a desirable learning outcome of introductory physics, where the math world and physical world meet. Physics Quantitative Literacy (PQL) is a set of interconnected skills and habits of mind that support quantitative reasoning about the physical world. The Physics Inventory of Quantitative Literacy (PIQL), which we are currently refining and validating, assesses students proportional reasoning, covariational reasoning, and reasoning with signed quantities in physics contexts. In this paper, we apply a Conceptual Blending Theory analysis of two exemplar PIQL items to demonstrate how we are using this theory to help develop an instrument that represents the kind of blended reasoning that characterizes expertise in physics. A Conceptual Blending Theory analysis allows for assessment of hierarchical partially correct reasoning patterns, and thereby holds potential to map the emergence of mathematical reasoning flexibility throughout the introductory physics sequence.
[ 73, 89, 97 ]
q-bio.CB q-bio.SC q-bio.TO
Impact of chronic fetal hypoxia and inflammation on cardiac pacemaker cell development. Chronic fetal hypoxia and infection are examples of adverse conditions during complicated pregnancy, which impact cardiac myogenesis and increase the lifetime risk of heart disease. However, the effects that chronic hypoxic or inflammatory environments exert on cardiac pacemaker cells are poorly understood. Here, we review the current evidence and novel avenues of bench to bed research in this field of perinatal cardiogenesis as well as its translational significance for early detection of future risk for cardiovascular disease.
[ 32, 20, 68 ]
math.NA cs.DM cs.DS cs.NA
Derivation and Analysis of Fast Bilinear Algorithms for Convolution. The prevalence of convolution in applications within signal processing, deep neural networks, and numerical solvers has motivated the development of numerous fast convolution algorithms. In many of these problems, convolution is performed on terabytes or petabytes of data, so even constant factors of improvement can significantly reduce the computation time. We leverage the formalism of bilinear algorithms to describe and analyze all of the most popular approaches. This unified lens permits us to study the relationship between different variants of convolution as well as to derive error bounds and analyze the cost of the various algorithms. We provide new derivations, which predominantly leverage matrix and tensor algebra, to describe the Winograd family of convolution algorithms as well as reductions between 1D and multidimensional convolution. We provide cost and error bounds as well as experimental numerical studies. Our experiments for two of these algorithms, the overlap add approach and Winograd convolution algorithm with polynomials of degree greater than one, show that fast convolution algorithms can rival the accuracy of the fast Fourier transform (FFT) without using complex arithmetic. These algorithms can be used for convolution problems with multidimensional inputs or for filters larger than size of four, extending the state of the art in Winograd based convolution algorithms.
[ 64, 10 ]
physics.soc-ph physics.ao-ph
Sustainable financing of permanent CO2 disposal through a Carbon Takeback Obligation. Unless there is immediate, unprecedented, reduction in global demand for carbon intensive energy and products, then capture and permanent storage of billions of tonnes of carbon dioxide (CO2) annually will be needed before mid century to meet Paris Agreement goals. Yet competition from cheaper, temporary, carbon storage means that permanent disposal remains starved of investment, currently representing about 0.1 of Energy and Industrial Process (EIP) emissions. This stored fraction must reach 100 to stop EIPs causing global warming. Here we show that a cost effective transition can occur by mandating an increasing stored fraction through a progressive Carbon Takeback Obligation (CTO) on fossil fuel producers and importers. Projected costs of storage to the consumer are lower than pricing carbon emissions in conventional 1.5 deg C scenarios until the 2040s, and comparable or lower thereafter. A CTO combined with measures to reduce CO2 production would deliver the lowest risk pathway to achieving net zero.
[ 0, 2 ]
cs.CV cs.AI
Synbols Probing Learning Algorithms with Synthetic Datasets. Progress in the field of machine learning has been fueled by the introduction of benchmark datasets pushing the limits of existing algorithms. Enabling the design of datasets to test specific properties and failure modes of learning algorithms is thus a problem of high interest, as it has a direct impact on innovation in the field. In this sense, we introduce Synbols Synthetic Symbols a tool for rapidly generating new datasets with a rich composition of latent features rendered in low resolution images. Synbols leverages the large amount of symbols available in the Unicode standard and the wide range of artistic font provided by the open font community. Our tool's high level interface provides a language for rapidly generating new distributions on the latent features, including various types of textures and occlusions. To showcase the versatility of Synbols, we use it to dissect the limitations and flaws in standard learning algorithms in various learning setups including supervised learning, active learning, out of distribution generalization, unsupervised representation learning, and object counting.
[ 0, 41 ]
cs.CV cs.RO
PCAN 3D Attention Map Learning Using Contextual Information for Point Cloud Based Retrieval. Point cloud based retrieval for place recognition is an emerging problem in vision field. The main challenge is how to find an efficient way to encode the local features into a discriminative global descriptor. In this paper, we propose a Point Contextual Attention Network (PCAN), which can predict the significance of each local point feature based on point context. Our network makes it possible to pay more attention to the task relevent features when aggregating local features. Experiments on various benchmark datasets show that the proposed network can provide outperformance than current state of the art approaches.
[ 8, 33 ]
physics.ins-det physics.optics
Comb mode resolved adaptive sampling terahertz dual comb spectroscopy with a free running single cavity fiber laser. Mode resolved dual comb spectroscopy (DCS) is an emerging spectroscopic tool with the potential to simultaneously achieve a broad spectral coverage and ultrahigh spectral resolution in terahertz (THz) spectroscopy. However, the need for two independently stabilized ultrafast lasers significantly hampers the potential application of DCS techniques. In this article, we demonstrate mode resolved DCS in the THz region based on a free running single cavity dual comb fiber laser with adaptive sampling. Low pressure spectroscopy of acetonitrile gas with absorption features approaching the Doppler limit is demonstrated by comb mode resolved measurements with a spectral sampling spacing of 48.8 MHz, a spectral resolution of less than 5 MHz and a signal to noise ratio of 50 dB. The successful demonstration of the proposed method clearly indicates the great potential for the realization of low complexity, MHz resolution THz spectroscopy instrumentation.
[ 41, 14 ]
cs.HC cs.RO
A VR System for Immersive Teleoperation and Live Exploration with a Mobile Robot. Applications like disaster management and industrial inspection often require experts to enter contaminated places. To circumvent the need for physical presence, it is desirable to generate a fully immersive individual live teleoperation experience. However, standard video based approaches suffer from a limited degree of immersion and situation awareness due to the restriction to the camera view, which impacts the navigation. In this paper, we present a novel VR based practical system for immersive robot teleoperation and scene exploration. While being operated through the scene, a robot captures RGB D data that is streamed to a SLAM based live multi client telepresence system. Here, a global 3D model of the already captured scene parts is reconstructed and streamed to the individual remote user clients where the rendering for e.g. head mounted display devices (HMDs) is performed. We introduce a novel lightweight robot client component which transmits robot specific data and enables a quick integration into existing robotic systems. This way, in contrast to first person exploration systems, the operators can explore and navigate in the remote site completely independent of the current position and view of the capturing robot, complementing traditional input devices for teleoperation. We provide a proof of concept implementation and demonstrate the capabilities as well as the performance of our system regarding interactive object measurements and bandwidth efficient data streaming and visualization. Furthermore, we show its benefits over purely video based teleoperation in a user study revealing a higher degree of situation awareness and a more precise navigation in challenging environments.
[ 46 ]
math.AG
Families of divisors. We establish a new moduli theory for divisors, that interpolates between the Hilbert scheme and the Cayley Chow variety. This completes the last step in the construction of a good moduli theory for stable pairs (X, Delta) .
[ 1 ]
cs.LG
Fraud Detection using Data Driven approach. The extensive use of the internet is continuously drifting businesses to incorporate their services in the online environment. One of the first spectrums to embrace this evolution was the banking sector. In fact, the first known online banking service came in 1980. It was deployed from a community bank located in Knoxville, called the United American Bank. Since then, internet banking has been offering ease and efficiency to costumers in completing their daily banking tasks. The ever increasing use of internet banking and a large number of online transactions increased fraudulent behavior also. As if fraud increase was not enough, the massive number of online transactions further increased the data complexity. Modern data sources are not only complex but generated at high speed and in real time as well. This presents a serious problem and a definite reason why more advanced solutions are desired to protect financial service companies and credit cardholders. Therefore, this research paper aims to construct an efficient fraud detection model which is adaptive to customer behavior changes and tends to decrease fraud manipulation, by detecting and filtering fraud in real time. In order to achieve this aim, a review of various methods is conducted, adding above a personal experience working in the Banking sector, specifically in the Fraud Detection office. Unlike the majority of reviewed methods, the proposed model in this research paper is able to detect fraud in the moment of occurrence using an incremental classifier. The evaluation of synthetic data, based on fraud scenarios selected in collaboration with domain experts that replicate typical, real world attacks, shows that this approach correctly ranks complex frauds. In particular, our proposal detects fraudulent behavior and anomalies with up to 97 detection rate while maintaining a satisfyingly low cost.
[ 0 ]
cs.CV
Cost Volume Pyramid Based Depth Inference for Multi View Stereo. We propose a cost volume based neural network for depth inference from multi view images. We demonstrate that building a cost volume pyramid in a coarse to fine manner instead of constructing a cost volume at a fixed resolution leads to a compact, lightweight network and allows us inferring high resolution depth maps to achieve better reconstruction results. To this end, we first build a cost volume based on uniform sampling of fronto parallel planes across the entire depth range at the coarsest resolution of an image. Then, given current depth estimate, we construct new cost volumes iteratively on the pixelwise depth residual to perform depth map refinement. While sharing similar insight with Point MVSNet as predicting and refining depth iteratively, we show that working on cost volume pyramid can lead to a more compact, yet efficient network structure compared with the Point MVSNet on 3D points. We further provide detailed analyses of the relation between (residual) depth sampling and image resolution, which serves as a principle for building compact cost volume pyramid. Experimental results on benchmark datasets show that our model can perform 6x faster and has similar performance as state of the art methods. Code is available at https github.com JiayuYANG CVP MVSNet
[ 1, 15 ]
cs.IT cs.LG math.IT
Neural Entropic Estimation A faster path to mutual information estimation. We point out a limitation of the mutual information neural estimation (MINE) where the network fails to learn at the initial training phase, leading to slow convergence in the number of training iterations. To solve this problem, we propose a faster method called the mutual information neural entropic estimation (MI NEE). Our solution first generalizes MINE to estimate the entropy using a custom reference distribution. The entropy estimate can then be used to estimate the mutual information. We argue that the seemingly redundant intermediate step of entropy estimation allows one to improve the convergence by an appropriate reference distribution. In particular, we show that MI NEE reduces to MINE in the special case when the reference distribution is the product of marginal distributions, but faster convergence is possible by choosing the uniform distribution as the reference distribution instead. Compared to the product of marginals, the uniform distribution introduces more samples in low density regions and fewer samples in high density regions, which appear to lead to an overall larger gradient for faster convergence.
[ 24, 42, 21 ]
math.MG math.CO math.DG
Discrete harmonic functions on infinite penny graphs. In this paper, we study discrete harmonic functions on infinite penny graphs. For an infinite penny graph with bounded facial degree, we prove that the volume doubling property and the Poincar 'e inequality hold, which yields the Harnack inequality for positive harmonic functions. Moreover, we prove that the space of polynomial growth harmonic functions, or ancient solutions of the heat equation, with bounded growth rate has finite dimensional property.
[ 39 ]
cs.CY
Reproducibility via Crowdsourced Reverse Engineering A Neural Network Case Study With DeepMind's Alpha Zero. The reproducibility of scientific findings are an important hallmark of quality and integrity in research. The scientific method requires hypotheses to be subjected to the most crucial tests, and for the results to be consistent across independent trials. Therefore, a publication is expected to provide sufficient information for an objective evaluation of its methods and claims. This is particularly true for research supported by public funds, where transparency of findings are a form of return on public investment. Unfortunately, many publications fall short of this mark for various reasons, including unavoidable ones such as intellectual property protection and national security of the entity creating those findings. This is a particularly important and documented problem in medical research, and in machine learning. Fortunately for those seeking to overcome these difficulties, the internet makes it easier to share experiments, and allows for crowd sourced reverse engineering. A case study of this capability in neural networks research is presented in this paper. The significant success of reverse engineering the important accomplishments of DeepMind's Alpha Zero exemplifies the leverage that can be achieved by a concerted effort to reproduce results.
[ 10 ]
physics.ao-ph
Trend Analysis of Meteorological Parameters, Tropospheric Refractivity, Equivalent Potential Temperature for a Pseudoadiabatic Process and Field Strength Variability, Using Mann Kendall Trend Test and Sens Estimate. Trend analysis of meteorological parameters (temperature, pressure, and relative humidity) as well as calculated refractivity, equivalent potential temperature (EPT) for a pseudo adiabatic process, and field strength in Calabar, Southern Nigeria has been analyzed using Mann Kendall trend test and Sens slope estimator. Data of the meteorological parameters were obtained from the Nigerian Meteorological Agency (NiMet) in Calabar for 14 years (2005 2018). Results show that the maximum and average temperature, atmospheric pressure, refractivity, EPT and field strength all exhibited a positive Kendall Z value with 2.52, 0.33, 3.83, 0.77, 0.44 and 3.18 respectively which indicated an increasing trend over time, with only maximum temperature, atmospheric pressure and field strength showing a significant increase at 5 (0.05) level of significance, since their calculated p values (0.012, 0.0001, and 0.001) were less than 0.05. The relative humidity and minimum ambient temperature showed a decrease in trend over time as they both had a negative Kendall Z values ( 0.11 and 1.09 respectively), however, together with the average ambient temperature and refractivity, their trend was not significant 5 level of significance since their calculated p values were all more than 0.05. Linear regression, correlation and partial differentiation showed that relative humidity has the most effect on the changes in seasonal refractivity and an indirect relationship with field strength variability. The relationship between EPT and refractivity has been discovered to be very strong and positive. Descriptive statistics has been used to portray the seasonal and annual trend of all parameters.
[ 9, 42 ]
math.AP math.DG
The Willmore Flow of Tori of Revolution. We study long time existence and asymptotic behavior for the L 2 gradient flow of the Willmore energy, under the condition that the initial datum is a torus of revolution. We show that if an initial datum has Willmore energy below 8 pi then the solution of the Willmore flow converges for t rightarrow infty to the Clifford Torus, possibly rescaled and translated. The energy threshold of 8 pi turns out to be optimal for such a convergence result. We give an application to the conformally constrained Willmore minimization problem.
[ 38 ]
math.NT
The first moment of primes in arithmetic progressions Beyond the Siegel Walfisz range. We investigate the first moment of primes in progressions sum substack q leq x N (q,a) 1 Big( psi(x q, a) frac x varphi(q) Big) as x, N to infty . We show unconditionally that, when a 1 , there is a significant bias towards negative values, uniformly for N leq rm e c sqrt log x . The proof combines recent results of the authors on the first moment and on the error term in the dispersion method. More generally, for a in mathbb Z setminus 0 we prove estimates that take into account the potential existence (or inexistence) of Landau Siegel zeros.
[ 56, 37 ]
cs.FL cs.LO
Extensions of omega Regular Languages. We consider extensions of monadic second order logic over omega words, which are obtained by adding one language that is not omega regular. We show that if the added language L has a neutral letter, then the resulting logic is necessarily undecidable. A corollary is that the omega regular languages are the only decidable Boolean closed full trio over omega words.
[ 3 ]
cs.CL
Fair Hate Speech Detection through Evaluation of Social Group Counterfactuals. Approaches for mitigating bias in supervised models are designed to reduce models' dependence on specific sensitive features of the input data, e.g., mentioned social groups. However, in the case of hate speech detection, it is not always desirable to equalize the effects of social groups because of their essential role in distinguishing outgroup derogatory hate, such that particular types of hateful rhetoric carry the intended meaning only when contextualized around certain social group tokens. Counterfactual token fairness for a mentioned social group evaluates the model's predictions as to whether they are the same for (a) the actual sentence and (b) a counterfactual instance, which is generated by changing the mentioned social group in the sentence. Our approach assures robust model predictions for counterfactuals that imply similar meaning as the actual sentence. To quantify the similarity of a sentence and its counterfactual, we compare their likelihood score calculated by generative language models. By equalizing model behaviors on each sentence and its counterfactuals, we mitigate bias in the proposed model while preserving the overall classification performance.
[ 0, 1, 41 ]
cs.CV cs.LG cs.RO
Deep Reinforcement Learning with Mixed Convolutional Network. Recent research has shown that map raw pixels from a single front facing camera directly to steering commands are surprisingly powerful. This paper presents a convolutional neural network (CNN) to playing the CarRacing v0 using imitation learning in OpenAI Gym. The dataset is generated by playing the game manually in Gym and used a data augmentation method to expand the dataset to 4 times larger than before. Also, we read the true speed, four ABS sensors, steering wheel position, and gyroscope for each image and designed a mixed model by combining the sensor input and image input. After training, this model can automatically detect the boundaries of road features and drive the robot like a human. By comparing with AlexNet and VGG16 using the average reward in CarRacing v0, our model wins the maximum overall system performance.
[ 17 ]
cs.DC
Distributed Approximation Algorithms for Steiner Tree in the mathcal CONGESTED mathcal CLIQUE . The emph Steiner tree problem is one of the fundamental and classical problems in combinatorial optimization. In this paper, we study this problem in the mathcal CONGESTED mathcal CLIQUE model of distributed computing and present two deterministic distributed approximation algorithms for the same. The first algorithm computes a Steiner tree in tilde O (n 1 3 ) rounds and tilde O (n 7 3 ) messages for a given connected undirected weighted graph of n nodes. Note here that tilde O ( cdot) notation hides polylogarithmic factors in n . The second one computes a Steiner tree in O(S log log n) rounds and O(S (n t) 2 n 2) messages, where S and t are the emph shortest path diameter and the number of emph terminal nodes respectively in the given input graph. Both the algorithms admit an approximation factor of 2(1 1 ell) , where ell is the number of terminal leaf nodes in the optimal Steiner tree. For graphs with S omega(n 1 3 log n) , the first algorithm exhibits better performance than the second one in terms of the round complexity. On the other hand, for graphs with S tilde o (n 1 3 ) , the second algorithm outperforms the first one in terms of the round complexity. In fact when S O( log log n) then the second algorithm admits a round complexity of O( log log n) and message complexity of tilde O (n 2) . To the best of our knowledge, this is the first work to study the Steiner tree problem in the mathcal CONGESTED mathcal CLIQUE model.
[ 3 ]
cs.CL
PMIndia A Collection of Parallel Corpora of Languages of India. Parallel text is required for building high quality machine translation (MT) systems, as well as for other multilingual NLP applications. For many South Asian languages, such data is in short supply. In this paper, we described a new publicly available corpus (PMIndia) consisting of parallel sentences which pair 13 major languages of India with English. The corpus includes up to 56000 sentences for each language pair. We explain how the corpus was constructed, including an assessment of two different automatic sentence alignment methods, and present some initial NMT results on the corpus.
[ 2 ]
cs.AI
Order Matters Generating Progressive Explanations for Planning Tasks in Human Robot Teaming. Prior work on generating explanations in a planning and decision making context has focused on providing the rationale behind an AI agent's decision making. While these methods provide the right explanations from the explainer's perspective, they fail to heed the cognitive requirement of understanding an explanation from the explainee's (the human's) perspective. In this work, we set out to address this issue by first considering the influence of information order in an explanation, or the progressiveness of explanations. Intuitively, progression builds later concepts on previous ones and is known to contribute to better learning. In this work, we aim to investigate similar effects during explanation generation when an explanation is broken into multiple parts that are communicated sequentially. The challenge here lies in modeling the humans' preferences for information order in receiving such explanations to assist understanding. Given this sequential process, a formulation based on goal based MDP for generating progressive explanations is presented. The reward function of this MDP is learned via inverse reinforcement learning based on explanations that are retrieved via human subject studies. We first evaluated our approach on a scavenger hunt domain to demonstrate its effectively in capturing the humans' preferences. Upon analyzing the results, it revealed something more fundamental the preferences arise strongly from both domain dependent and independence features. The correlation with domain independent features pushed us to verify this result further in an escape room domain. Results confirmed our hypothesis that the process of understanding an explanation was a dynamic process. The human preference that reflected this aspect corresponded exactly to the progression for knowledge assimilation hidden deeper in our cognitive process.
[ 25, 23 ]
math.PR math.FA
Non local Markovian symmetric forms on infinite dimensional spaces. General theorems on the closability and quasi regularity of non local Markovian symmetric forms on probability spaces (S, cal B (S), mu) , with S Fr 'e chet spaces such that S subset mathbb R mathbb N , cal B (S) is the Borel sigma field of S , and mu is a Borel probability measure on S , are introduced. Firstly, a family of non local Markovian symmetric forms cal E ( alpha) , 0 alpha 2 , acting in each given L 2(S mu) is defined, the index alpha characterizing the order of the non locality. Then, it is shown that all the forms cal E ( alpha) defined on bigcup n in mathbb N C infty 0( mathbb R n) are closable in L 2(S mu) . Moreover, sufficient conditions under which the closure of the closable forms, that are Dirichlet forms, become strictly quasi regular, are given. Finally, an existence theorem for Hunt processes properly associated to the Dirichlet forms is given. The application of the above theorems to the problem of stochastic quantizations of Euclidean Phi 4 d fields, for d 2, 3 , by means of these Hunt processes is indicated.
[ 48, 62 ]
cs.GT cs.CC
On the Computational Complexity of Decision Problems about Multi Player Nash Equilibria. We study the computational complexity of decision problems about Nash equilibria in m player games. Several such problems have recently been shown to be computationally equivalent to the decision problem for the existential theory of the reals, or stated in terms of complexity classes, exists mathbb R complete, when m geq 3 . We show that, unless they turn into trivial problems, they are exists mathbb R hard even for 3 player zero sum games. We also obtain new results about several other decision problems. We show that when m geq 3 the problems of deciding if a game has a Pareto optimal Nash equilibrium or deciding if a game has a strong Nash equilibrium are exists mathbb R complete. The latter result rectifies a previous claim of NP completeness in the literature. We show that deciding if a game has an irrational valued Nash equilibrium is exists mathbb R hard, answering a question of Bil o and Mavronicolas, and address also the computational complexity of deciding if a game has a rational valued Nash equilibrium. These results also hold for 3 player zero sum games. Our proof methodology applies to corresponding decision problems about symmetric Nash equilibria in symmetric games as well, and in particular our new results carry over to the symmetric setting. Finally we show that deciding whether a symmetric m player games has a non symmetric Nash equilibrium is exists mathbb R complete when m geq 3 , answering a question of Garg, Mehta, Vazirani, and Yazdanbod.
[ 0 ]
cs.CV
Towards Fairer Datasets Filtering and Balancing the Distribution of the People Subtree in the ImageNet Hierarchy. Computer vision technology is being used by many but remains representative of only a few. People have reported misbehavior of computer vision models, including offensive prediction results and lower performance for underrepresented groups. Current computer vision models are typically developed using datasets consisting of manually annotated images or videos the data and label distributions in these datasets are critical to the models' behavior. In this paper, we examine ImageNet, a large scale ontology of images that has spurred the development of many modern computer vision methods. We consider three key factors within the "person" subtree of ImageNet that may lead to problematic behavior in downstream computer vision technology (1) the stagnant concept vocabulary of WordNet, (2) the attempt at exhaustive illustration of all categories with images, and (3) the inequality of representation in the images within concepts. We seek to illuminate the root causes of these concerns and take the first steps to mitigate them constructively.
[ 46, 38 ]
math.NT math.AG
Exceptional jumps of Picard ranks of reductions of K3 surfaces over number fields. Given a K3 surface X over a number field K with potentially good reduction everywhere, we prove that the set of primes of K where the geometric Picard rank jumps is infinite. As a corollary, we prove that either X overline K has infinitely many rational curves or X has infinitely many unirational specializations. Our result on Picard ranks is a special case of more general results on exceptional classes for K3 type motives associated to GSpin Shimura varieties. These general results have several other applications. For instance, we prove that an abelian surface over a number field K with potentially good reduction everywhere is isogenous to a product of elliptic curves modulo infinitely many primes of K .
[ 32 ]
cs.DS
Online Virtual Machine Allocation with Predictions. The cloud computing industry has grown rapidly over the last decade, and with this growth there is a significant increase in demand for compute resources. Demand is manifested in the form of Virtual Machine (VM) requests, which need to be assigned to physical machines in a way that minimizes resource fragmentation and efficiently utilizes the available machines. This problem can be modeled as a dynamic version of the bin packing problem with the objective of minimizing the total usage time of the bins (physical machines). Earlier works on dynamic bin packing assumed that no knowledge is available to the scheduler and later works studied models in which lifetime duration of each "item" (VM in our context) is available to the scheduler. This extra information was shown to improve exponentially the achievable competitive ratio. Motivated by advances in Machine Learning that provide good estimates of workload characteristics, this paper studies the effect of having extra information regarding future (total) demand. In the cloud context, since demand is an aggregate over many VM requests, it can be predicted with high accuracy (e.g., using historical data). We show that the competitive factor can be dramatically improved by using this additional information in some cases, we achieve constant competitiveness, or even a competitive factor that approaches 1. Along the way, we design new offline algorithms with improved approximation ratios for the dynamic bin packing problem.
[ 51 ]
physics.bio-ph
Covariance Distributions in Single Particle Tracking. Several recent experiments, including our own in the fission yeast, S. pombe, have characterized the motions of gene loci within living nuclei by measuring the locus position over time, then proceeding to obtain the statistical properties of this motion. To address the question of whether a population of single particle tracks, obtained from many different cells, corresponds to a single mode of diffusion, we derive theoretical equations describing the probability distribution of the displacement covariance, assuming the displacement is a zero mean multivariate Gaussian random variable. We also determine the corresponding theoretical means, variances, and third central moments. Bolstering the theory is good agreement between its predictions and the results obtained for various simulated and measured data sets, including simulated particle trajectories of simple and anomalous diffusion, and the measured trajectories of an optically trapped bead in water, and in a viscoelastic solution. We also show that, for sufficiently long tracks, each covariance distribution in these examples is well described by a skew normal distribution with mean, variance, and skewness given by theory. For experimental S. pombe gene locus data, however, we find that the first two covariance distributions are wider than predicted, although the third and subsequent covariances are well described by theory. This suggests that the origin of the theory experiment discrepancy is associated with localization noise, which influences only the first two covariances. Thus, we hypothesize that the discrepancy is caused by locus to locus heterogeneity in the localization noise. Further simulations reveal excess covariance widths can be largely recreated on the basis of heterogeneous noise. We conclude that the motion of gene loci in fission yeast is consistent with a single mode of diffusion.
[ 9, 20 ]
math.NA cs.NA math.AP
Manifold Learning and Nonlinear Homogenization. We describe an efficient domain decomposition based framework for nonlinear multiscale PDE problems. The framework is inspired by manifold learning techniques and exploits the tangent spaces spanned by the nearest neighbors to compress local solution manifolds. Our framework is applied to a semilinear elliptic equation with oscillatory media and a nonlinear radiative transfer equation in both cases, significant improvements in efficacy are observed. This new method does not rely on detailed analytical understanding of the multiscale PDEs, such as their asymptotic limits, and thus is more versatile for general multiscale problems.
[ 18, 21 ]
math.GT math.CO
Lower bounds for regular genus and gem complexity of PL 4 manifolds with boundary. Let M be a connected compact PL 4 manifold with boundary. In this article, we have given several lower bounds for regular genus and gem complexity of the manifold M . In particular, we have proved that if M is a connected compact 4 manifold with h boundary components then its gem complexity mathit k (M) satisfies the following inequalities mathit k (M) geq 3 chi(M) 7m 7h 10 mbox and mathit k (M) geq mathit k ( partial M) 3 chi(M) 4m 6h 9, and its regular genus mathcal G (M) satisfies the following inequalities mathcal G (M) geq 2 chi(M) 3m 2h 4 mbox and mathcal G (M) geq mathcal G ( partial M) 2 chi(M) 2m 2h 4, where m is the rank of the fundamental group of the manifold M . These lower bounds enable to strictly improve previously known estimations for regular genus and gem complexity of a PL 4 manifold with boundary. Further, the sharpness of these bounds has also been shown for a large class of PL 4 manifolds with boundary.
[ 8 ]
physics.optics
Analytical vs. Numerical Langevin Description of Noise in Small Lasers. We compare the analytical and numerical predictions of noise in nano and microcavity lasers obtained from a rate equation model with stochastic Langevin noise. Strong discrepancies are found between the two approaches and these are critically analyzed and explained on the basis of general considerations and through the comparison to the numerical predictions of a Stochastic Laser Simulator. While the analytical calculations give reliable redictions, the numerical results are entirely incorrect thus unsuitable for predicting the dynamics and statistical properties of small lasers.
[ 42 ]
math.DG
Non tame Morse Smale flows and odd Chern Weil theory. Using a certain well posed ODE problem introduced by Shilnikov in the sixties, G. Minervini proved in his PhD thesis 17 , among other things, the Harvey Lawson Diagonal Theorem but without the restrictive tameness condition for Morse flows. Here we combine the same techniques with the ideas of Latschev in order to construct local resolutions for the flow of the graph of a section of a fiber bundle. This is endowed with a vertical vector field which is horizontally constant and Morse Smale in every fiber. The resolution allows the removal of the tameness hypothesis from the homotopy formula in 3 . We give one finite and one infinite dimensional application. For that end, we introduce closed smooth forms of odd degree associated to any triple (E U nabla) composed of a hermitian vector bundle, unitary endomorphism and metric compatible connection.
[ 65, 17, 1 ]
cs.DC cs.LG cs.NI
A Computational Approach to Packet Classification. Multi field packet classification is a crucial component in modern software defined data center networks. To achieve high throughput and low latency, state of the art algorithms strive to fit the rule lookup data structures into on die caches however, they do not scale well with the number of rules. We present a novel approach, NuevoMatch, which improves the memory scaling of existing methods. A new data structure, Range Query Recursive Model Index (RQ RMI), is the key component that enables NuevoMatch to replace most of the accesses to main memory with model inference computations. We describe an efficient training algorithm that guarantees the correctness of the RQ RMI based classification. The use of RQ RMI allows the rules to be compressed into model weights that fit into the hardware cache. Further, it takes advantage of the growing support for fast neural network processing in modern CPUs, such as wide vector instructions, achieving a rate of tens of nanoseconds per lookup. Our evaluation using 500K multi field rules from the standard ClassBench benchmark shows a geometric mean compression factor of 4.9x, 8x, and 82x, and average performance improvement of 2.4x, 2.6x, and 1.6x in throughput compared to CutSplit, NeuroCuts, and TupleMerge, all state of the art algorithms.
[ 6, 7, 10, 55, 93 ]
physics.class-ph physics.ao-ph physics.comp-ph physics.flu-dyn physics.geo-ph
Long wave interaction with a partially immersed body. Part I Mathematical models. In the present article we consider the problem of wave interaction with a partially immersed, but floating body. We assume that the motion of the body is prescribed. The general mathematical formulation for this problem is presented in the framework of a hierarchy of mathematical models. Namely, in this first part we formulate the problem at every hierarchical level. The special attention is payed to fully nonlinear and weakly dispersive models since they are most likely to be used in practice. For this model we have to consider separately the inner (under the body) and outer domains. Various approached to the gluing of solutions at the boundary is discussed as well. We propose several strategies which ensure the global conservation or continuity of some important physical quantities.
[ 1, 3, 92 ]
cs.CL cs.LG q-bio.BM
BERTology Meets Biology Interpreting Attention in Protein Language Models. Transformer architectures have proven to learn useful representations for protein classification and generation tasks. However, these representations present challenges in interpretability. In this work, we demonstrate a set of methods for analyzing protein Transformer models through the lens of attention. We show that attention (1) captures the folding structure of proteins, connecting amino acids that are far apart in the underlying sequence, but spatially close in the three dimensional structure, (2) targets binding sites, a key functional component of proteins, and (3) focuses on progressively more complex biophysical properties with increasing layer depth. We find this behavior to be consistent across three Transformer architectures (BERT, ALBERT, XLNet) and two distinct protein datasets. We also present a three dimensional visualization of the interaction between attention and protein structure. Code for visualization and analysis is available at https github.com salesforce provis.
[ 25 ]
math.PR
On the local limit theorems for psi mixing Markov chains. In this paper we investigate the local limit theorem for additive functionals of a nonstationary Markov chain with finite or infinite second moment. The moment conditions are imposed on the individual summands and the weak dependence structure is expressed in terms of some uniformly mixing coefficients.
[ 11 ]
physics.app-ph
Material and device characterization of Type II InAS GaSb superlattice infrared detectors. This work investigates midwave infrared Type II InAs GaSb superlattice (SL) grown by molecular beam epitaxy on GaSb substrate. In order to compensate the natural tensile strain of the InAs layers, two different shutter sequences have been explored during the growth. The first one consists of growing an intentional InSb layer at both interfaces (namely GaSb on InAs and InAs on GaSb interfaces) by migration enhanced epitaxy while the second uses the antimony for arsenic exchange to promote an InSb like interface at the GaSb on InAs interface. SLs obtained via both methods are compared in terms of structural, morphological and optical properties by means of high resolution x ray diffraction, atomic force microscopy and photoluminescence spectroscopy. By using the second method, we obtained a nearly strain compensated SL on GaSb with a full width at half maximum of 56 arcsec for the first order SL satellite peak. Relatively smooth surface has been achieved with a root mean square value of about 0.19 nm on a 2 mu m x 2 mu m scan area. Finally, a p i n device structure having a cut off wavelength of 5.15 mu m at 77K has been demonstrated with a dark current level of 8.3 10 8 A cm 2 at 50 mV and a residual carrier concentration of 9.7 10 14 cm 3 , comparable to the state of the art.
[ 13 ]
math.DS
Comparison and Simplicity of Commutator Subgroups of Full Groups. We show that for a minimal, second countable, locally compact Hausdorff 'etale groupoid whose unit space is homeomorphic to the Cantor set, if the groupoid has comparison then the commutator subgroup of its full group is simple. This generalizes a result of Bezuglyi and Medynets for Cantor minimal systems and complements Matui's results for topological full groups.
[ 29 ]
math.GR
Short Presentations for Aut n and Out n . This paper uses the methods from arXiv 0804.1396 to give 'short' presentations for Aut (F n) , the special automorphism group of the free group of rank n , and Out (F n) the special outer automorphism group of the free group of rank n .
[ 32, 17 ]
cs.DS cs.DC
Improved Local Computation Algorithm for Set Cover via Sparsification. We design a Local Computation Algorithm (LCA) for the set cover problem. Given a set system where each set has size at most s and each element is contained in at most t sets, the algorithm reports whether a given set is in some fixed set cover whose expected size is O( log s ) times the minimum fractional set cover value. Our algorithm requires s O( log s ) t O( log s cdot ( log log s log log t )) queries. This result improves upon the application of the reduction of Parnas and Ron, TCS'07 on the result of Kuhn et al., SODA'06 , which leads to a query complexity of (st) O( log s cdot log t ) . To obtain this result, we design a parallel set cover algorithm that admits an efficient simulation in the LCA model by using a sparsification technique introduced in Ghaffari and Uitto, SODA'19 for the maximal independent set problem. The parallel algorithm adds a random subset of the sets to the solution in a style similar to the PRAM algorithm of Berger et al., FOCS'89 . However, our algorithm differs in the way that it never revokes its decisions, which results in a fewer number of adaptive rounds. This requires a novel approximation analysis which might be of independent interest.
[ 67, 36 ]
physics.atom-ph physics.chem-ph
Second Scale Coherence Measured at the Quantum Projection Noise Limit with Hundreds of Molecular Ions. Cold molecules provide an excellent platform for quantum information, cold chemistry, and precision measurement. Certain molecules have enhanced sensitivity to beyond Standard Model physics, such as the electron's electric dipole moment ( e EDM). Molecular ions are easily trappable and are therefore particularly attractive for precision measurements where sensitivity scales with interrogation time. Here, we demonstrate a spin precession measurement with second scale coherence at the quantum projection noise (QPN) limit with hundreds of trapped molecular ions, chosen for their sensitivity to the e EDM rather than their amenability to state control and readout. Orientation resolved resonant photodissociation allows us to simultaneously measure two quantum states with opposite e EDM sensitivity, reaching the QPN limit and fully exploiting the high count rate and long coherence.
[ 81, 20 ]
cs.MS cs.NA math.NA
PCPATCH software for the topological construction of multigrid relaxation methods. Effective relaxation methods are necessary for good multigrid convergence. For many equations, standard Jacobi and Gau ss Seidel are inadequate, and more sophisticated space decompositions are required examples include problems with semidefinite terms or saddle point structure. In this paper we present a unifying software abstraction, PCPATCH, for the topological construction of space decompositions for multigrid relaxation methods. Space decompositions are specified by collecting topological entities in a mesh (such as all vertices or faces) and applying a construction rule (such as taking all degrees of freedom in the cells around each entity). The software is implemented in PETSc and facilitates the elegant expression of a wide range of schemes merely by varying solver options at runtime. In turn, this allows for the very rapid development of fast solvers for difficult problems.
[ 8, 11 ]
physics.app-ph physics.optics
Coaxial nanowires as plasmon mediated remote nanosensors. This study reports on the plasmon mediated remote Raman sensing promoted by specially designed coaxial nanowires. This unusual geometry for Raman study is based on the separation, by several micrometres, of the excitation laser spot, on one tip of the nanowire, and the Raman detection at the other tip. The very weak efficiency of Raman emission makes it challenging in a remote configuration. For the proof of concept, we designed coaxial nanowires consisting in a gold core to propagate the surface plasmon polaritons and a Raman emitting shell of poly(3,4 ethylene dioxythiophene). The success of the fabrication was demonstrated by correlating, for the same single nanowire, a morphological analysis by electron microscopy and Raman spectroscopy analysis. Importantly for probing remote Raman effect, the original hard template based process allows to control the location of the polymer shell all along the nanowire, or only close to one or the two nanowire tips. Such all in one single nanowires could have applications in the remote detection of photo degradable substances and for exploring 1D nanosources for integrated photonic and plasmonic systems.
[ 94 ]
physics.ed-ph
Case of two electrostatics problems Can providing a diagram adversely impact introductory physics students' problem solving performance?. Drawing appropriate diagrams is a useful problem solving heuristic that can transform a problem into a representation that is easier to exploit for solving it. One major focus while helping introductory physics students learn effective problem solving is to help them understand that drawing diagrams can facilitate problem solution. We conducted an investigation in which two different interventions were implemented during recitation quizzes in a large enrollment algebra based introductory physics course. Students were either (i) asked to solve problems in which the diagrams were drawn for them or (ii) explicitly told to draw a diagram. A comparison group was not given any instruction regarding diagrams. We developed rubrics to score the problem solving performance of students in different intervention groups and investigated ten problems. We found that students who were provided diagrams never performed better and actually performed worse than the other students on three problems, one involving standing sound waves in a tube (discussed elsewhere) and two problems in electricity which we focus on here. These two problems were the only problems in electricity that involved considerations of initial and final conditions, which may partly account for why students provided with diagrams performed significantly worse than students who were not provided with diagrams. In order to explore potential reasons for this finding, we conducted interviews with students and found that some students provided with diagrams may have spent less time on the conceptual analysis and planning stage of the problem solving process. In particular, those provided with the diagram were more likely to jump into the implementation stage of problem solving early without fully analyzing and understanding the problem, which can increase the likelihood of mistakes in solutions.
[ 14 ]
cs.HC
Tiger Wearable Glasses for the 20 20 20 Rule to Alleviate Computer Vision Syndrome. We propose Tiger, an eyewear system for helping users follow the 20 20 20 rule to alleviate the Computer Vision Syndrome symptoms. It monitors user's screen viewing activities and provides real time feedback to help users follow the rule. For accurate screen viewing detection, we devise a light weight multi sensory fusion approach with three sensing modalities, color, IMU, and lidar. We also design the real time feedback to effectively lead users to follow the rule. Our evaluation shows that Tiger accurately detects screen viewing events, and is robust to the differences in screen types, contents, and ambient light. Our user study shows positive perception of Tiger regarding its usefulness, acceptance, and real time feedback.
[ 21 ]
math.CO
On Generalised Petersen Graphs of Girth 7 that have Cop Number 4. We show that if n 7k i with i in 1,2,3 then the cop number of the generalised Petersen graph GP(n,k) is 4 , with some small previously known exceptions. It was previously proved by Ball et al. (2015) that the cop number of any generalised Petersen graph is at most 4 . The results in this paper explain all of the known generalised Petersen graphs that actually have cop number 4 but were not previously explained by Morris et al. in a recent preprint, and places them in the context of infinite families. (More precisely, the preprint by Morris et al. explains all known generalised Petersen graphs with cop number 4 and girth 8 , while this paper explains those that have girth 7 .)
[ 1, 3 ]
cs.CL cs.LG
Investigating Novel Verb Learning in BERT Selectional Preference Classes and Alternation Based Syntactic Generalization. Previous studies investigating the syntactic abilities of deep learning models have not targeted the relationship between the strength of the grammatical generalization and the amount of evidence to which the model is exposed during training. We address this issue by deploying a novel word learning paradigm to test BERT's few shot learning capabilities for two aspects of English verbs alternations and classes of selectional preferences. For the former, we fine tune BERT on a single frame in a verbal alternation pair and ask whether the model expects the novel verb to occur in its sister frame. For the latter, we fine tune BERT on an incomplete selectional network of verbal objects and ask whether it expects unattested but plausible verb object pairs. We find that BERT makes robust grammatical generalizations after just one or two instances of a novel word in fine tuning. For the verbal alternation tests, we find that the model displays behavior that is consistent with a transitivity bias verbs seen few times are expected to take direct objects, but verbs seen with direct objects are not expected to occur intransitively.
[ 77 ]
math.LO
L Topology via Generalised Geometric Logic. This paper introduces a notion of generalised geometric logic. Connections of generalised geometric logic with L topological system and L topological space are established.
[ 64, 59, 31 ]
physics.soc-ph cs.SI q-bio.PE
Epidemic Spreading and Equilibrium Social Distancing in Heterogeneous Networks. We study a multi type SIR epidemic process among a heterogeneous population that interacts through a network. When we base social contact on a random graph with given vertex degrees, we give limit theorems on the fraction of infected individuals. For a given social distancing individual strategies, we establish the epidemic reproduction number R 0 which can be used to identify network vulnerability and inform vaccination policies. In the second part of the paper we study the equilibrium of the social distancing game, in which individuals choose their social distancing level according to an anticipated global infection rate, which then must equal the actual infection rate following their choices. We give conditions for the existence and uniqueness of equilibrium. For the case of random regular graphs, we show that voluntary social distancing will always be socially sub optimal.
[ 35, 29 ]
math.GR math.RA
The Modular Isomorphism Problem for small groups revisiting Eick's algorithm. We study the Modular Isomorphism Problem (MIP) for groups of small order based on an improvement of an algorithm described by B. Eick. Our improvement allows to determine quotients I(kG) I(kG) m of the augmentation ideal without first computing the full augmentation ideal I(kG) . It allows us to verify that the MIP has a positive answer for groups of order 3 7 and to significantly reduce the cases that need to be checked for groups of order 5 6 . We further provide a proof for an observation of Bagi 'nski and provide a negative answer to a question of Bleher, Kimmerle, Roggenkamp and Wursthorn.
[ 0, 41 ]
cs.RO cs.CV
Crop Height and Plot Estimation for Phenotyping from Unmanned Aerial Vehicles using 3D LiDAR. We present techniques to measure crop heights using a 3D Light Detection and Ranging (LiDAR) sensor mounted on an Unmanned Aerial Vehicle (UAV). Knowing the height of plants is crucial to monitor their overall health and growth cycles, especially for high throughput plant phenotyping. We present a methodology for extracting plant heights from 3D LiDAR point clouds, specifically focusing on plot based phenotyping environments. We also present a toolchain that can be used to create phenotyping farms for use in Gazebo simulations. The tool creates a randomized farm with realistic 3D plant and terrain models. We conducted a series of simulations and hardware experiments in controlled and natural settings. Our algorithm was able to estimate the plant heights in a field with 112 plots with a root mean square error (RMSE) of 6.1 cm. This is the first such dataset for 3D LiDAR from an airborne robot over a wheat field. The developed simulation toolchain, algorithmic implementation, and datasets can be found on the GitHub repository located at https github.com hsd1121 PointCloudProcessing.
[ 0 ]
cs.CV
Non negative representation based discriminative dictionary learning for face recognition. In this paper, we propose a non negative representation based discriminative dictionary learning algorithm (NRDL) for multicategory face classification. In contrast to traditional dictionary learning methods, NRDL investigates the use of non negative representation (NR), which contributes to learning discriminative dictionary atoms. In order to make the learned dictionary more suitable for classification, NRDL seamlessly incorporates nonnegative representation constraint, discriminative dictionary learning and linear classifier training into a unified model. Specifically, NRDL introduces a positive constraint on representation matrix to find distinct atoms from heterogeneous training samples, which results in sparse and discriminative representation. Moreover, a discriminative dictionary encouraging function is proposed to enhance the uniqueness of class specific sub dictionaries. Meanwhile, an inter class incoherence constraint and a compact graph based regularization term are constructed to respectively improve the discriminability of learned classifier. Experimental results on several benchmark face data sets verify the advantages of our NRDL algorithm over the state of the art dictionary learning methods.
[ 18 ]
math.GT
The third term in lens surgery polynomials. It is well known that the second coefficient of the Alexander polynomial of any lens space knot in S 3 is 1 . We show that the non zero third coefficient condition of the Alexander polynomial of a lens space knot K in S 3 confines the surgery to the one realized by the (2,2g 1) torus knot, where g is the genus of K . In particular, such a lens surgery polynomial coincides with Delta T(2,2g 1) (t) .
[ 41, 2, 1 ]
cs.RO cs.AI cs.LG
Human Robot Collaboration via Deep Reinforcement Learning of Real World Interactions. We present a robotic setup for real world testing and evaluation of human robot and human human collaborative learning. Leveraging the sample efficiency of the Soft Actor Critic algorithm, we have implemented a robotic platform able to learn a non trivial collaborative task with a human partner, without pre training in simulation, and using only 30 minutes of real world interactions. This enables us to study Human Robot and Human Human collaborative learning through real world interactions. We present preliminary results, showing that state of the art deep learning methods can take human robot collaborative learning a step closer to that of humans interacting with each other.
[ 8, 11 ]
physics.optics physics.app-ph
On chip spectrometer using stratified waveguides filters. We present an ultra compact single shot spectrometer on silicon platform with broad operation bandwidth and high resolution. It consists of 32 stratified waveguide filters (SWFs) with diverse transmission spectra for sampling the unknown spectrum of the input signal and a specially designed ultra compact structure for splitting the incident signal into 32 filters with low imbalance. Each SWF has a footprint less than 1um x 30um, while the 1x32 splitter and 32 filters in total occupy an area of about 35um x 260um, which to the best of our knowledge, is the smallest footprint spectrometer realized on silicon photonic platform. Experimental characteristics of the fabricated spectrometer demonstrate a broad operating bandwidth of 180nm centered at 1550nm and narrowband peaks with 0.45nm Full Width Half Maximum (FWHM) can be clearly resolved. This concept can also be implemented using other material platforms for operation in optical spectral bands of interest for various applications.
[ 49 ]
math.OC
General Asymptotic Regional Gradient Observer. The main purpose of this paper is to study and characterize the existing of general asymptotic regional gradient observer which observe the current gradient state of the original system in connection with gradient strategic sensors. Thus, we give an approach based to Luenberger observer theory of linear distributed parameter systems which is enabled to determinate asymptotically regional gradient estimator of current gradient system state. More precisely, under which condition the notion of asymptotic regional gradient observability can be achieved. Furthermore, we show that the measurement structures allows the existence of general asymptotic regional gradient observer and we give a sufficient condition for such asymptotic regional gradient observer in general case. We also show that, there exists a dynamical system for the considered system is not general asymptotic gradient observer in the usual sense, but it may be general asymptotic regional gradient observer. Then, for this purpose we present various results related to different types of sensor structures, domains and boundary conditions in two dimensional distributed diffusion systems
[ 41, 14, 39 ]
cs.CY cs.HC cs.RO
Migratable AI Effect of identity and information migration on users perception of conversational AI agents. Conversational AI agents are proliferating, embodying a range of devices such as smart speakers, smart displays, robots, cars, and more. We can envision a future where a personal conversational agent could migrate across different form factors and environments to always accompany and assist its user to support a far more continuous, personalized, and collaborative experience. This opens the question of what properties of a conversational AI agent migrates across forms, and how it would impact user perception. To explore this, we developed a Migratable AI system where a user's information and or the agent's identity can be preserved as it migrates across form factors to help its user with a task. We designed a 2x2 between subjects study to explore the effects of information migration and identity migration on user perceptions of trust, competence, likeability, and social presence. Our results suggest that identity migration had a positive effect on trust, competence, and social presence, while information migration had a positive effect on trust, competence, and likeability. Overall, users report the highest trust, competence, likeability, and social presence towards the conversational agent when both identity and information were migrated across embodiments.
[ 64 ]
physics.soc-ph
Mobility patterns of the Portuguese population during the COVID 19 pandemic. SARS CoV 2 emerged in late 2019. Since then, it has spread to several countries, becoming classified as a pandemic. So far, there is no definitive treatment or vaccine, so the best solution is to prevent transmission between individuals through social distancing. However, it is difficult to measure the effectiveness of these distance measures. Therefore, this study uses data from Google COVID 19 Community Mobility Reports to try to understand the mobility patterns of the Portuguese population during the COVID 19 pandemic. In this study, the Rt value was modeled for Portugal. Also, the changepoint was calculated for the population mobility patterns. Thus, the change in the mobility pattern was used to understand the impact of social distance measures on the dissemination of COVID 19. As a result, it can be stated that the initial Rt value in Portugal was very close to 3, falling to values close to 1 after 25 days. Social isolation measures were adopted quickly. Furthermore, it was observed that public transport was avoided during the pandemic. Finally, until the emergence of a vaccine or an effective treatment, this is the new normal, and it must be understood that new patterns of mobility, social interaction, and hygiene must be adapted to this reality.
[ 56, 57 ]
cs.PL cs.LO
RHLE Modular Deductive Verification of Relational forall exists Properties. Relational program logics are used to prove that a desired relationship holds between the execution of multiple programs. Existing relational program logics have focused on verifying that all runs of a collection of programs do not fall outside a desired set of behaviors. Several important relational properties, including refinement and noninterference, do not fit into this category, as they require the existence of specific desirable executions. This paper presents RHLE, a logic for verifying a class of relational properties which we term forall exists properties. forall exists properties assert that for all executions of a collection of programs, there exist executions of another set of programs exhibiting some intended behavior. Importantly, RHLE can reason modularly about programs which make library calls, ensuring that forall exists properties are preserved when the programs are linked with any valid implementation of the library. To achieve this, we develop a novel form of function specification that requires the existence of certain behaviors in valid implementations. We have built a tool based on RHLE which we use to verify a diverse set of relational properties drawn from the literature, including refinement and generalized noninterference.
[ 8, 13 ]
math.DS physics.optics
The limits of sustained self excitation and stable periodic pulse trains in the Yamada model with delayed optical feedback. We consider the Yamada model for an excitable or self pulsating laser with saturable absorber, and study the effects of delayed optical self feedback in the excitable case. More specifically, we are concerned with the generation of stable periodic pulse trains via repeated self excitation after passage through the delayed feedback loop, as well as their bifurcations. We show that onset and termination of such pulse trains correspond to the simultaneous bifurcation of countably many fold periodic orbits with infinite period in this delay differential equation. We employ numerical continuation and the concept of reappearance of periodic solutions to show that these bifurcations coincide with codimension two points along families of connecting orbits and fold periodic orbits in a related advanced differential equation. These points include heteroclinic connections between steady states, as well as homoclinic bifurcations with non hyperbolic equilibria. Tracking these codimension two points in parameter space reveals the critical parameter values for the existence of periodic pulse trains. We use the recently developed theory of temporal dissipative solitons to infer necessary conditions for the stability of such pulse trains.
[ 12, 76 ]
math.CA math.SP
The fundamental gap for a one dimensional Schr "odinger operator with Robin boundary conditions. For Schr "odinger operators on an interval with either convex or symmetric single well potentials, and Robin or Neumann boundary conditions, the gap between the two lowest eigenvalues is minimised when the potential is constant. We also have results for the p Laplacian.
[ 25, 23 ]
math.PR math.FA
Linear Dimension Reduction Approximately Preserving a Function of the 1 Norm. For any finite point set in D dimensional space equipped with the 1 norm, we present random linear embeddings to k dimensional space, with a new metric, having the following properties. For any pair of points from the point set that are not too close, the distance between their images is a strictly concave increasing function of their original distance, up to multiplicative error. The target dimension k need only be quadratic in the logarithm of the size of the point set to ensure the result holds with high probability. The linear embeddings are random matrices composed of standard Cauchy random variables, and the proofs rely on Chernoff bounds for sums of iid random variables. The new metric is translation invariant, but is not induced by a norm.
[ 36 ]
physics.chem-ph
A route to improving RPA excitation energies through its connection to equation of motion coupled cluster theory. We revisit the connection between equation of motion coupled cluster (EOM CC) and random phase approximation (RPA) explored recently by Berkelbach J. Chem. Phys. 149, 041103 (2018) and unify various methodological aspects of these diverse treatment of ground and excited states. The identity of RPA and EOM CC based on the ring coupled cluster doubles is established with numerical results which was proved previously on theoretical grounds. We then introduce new approximations in EOM CC and RPA family of methods, assess their numerical performance and explore a way to reap the benefits of such a connection to improve on excitation energies. Our results suggest that addition of perturbative corrections to account for double excitations and missing exchange effects could result in significantly improved estimates.
[ 1, 19 ]
cs.LG cs.CR
Evaluation of Federated Learning in Phishing Email Detection. The use of Artificial Intelligence (AI) to detect phishing emails is primarily dependent on large scale centralized datasets, which opens it up to a myriad of privacy, trust, and legal issues. Moreover, organizations are loathed to share emails, given the risk of leakage of commercially sensitive information. So, it is uncommon to obtain sufficient emails to train a global AI model efficiently. Accordingly, privacy preserving distributed and collaborative machine learning, particularly Federated Learning (FL), is a desideratum. Already prevalent in the healthcare sector, questions remain regarding the effectiveness and efficacy of FL based phishing detection within the context of multi organization collaborations. To the best of our knowledge, the work herein is the first to investigate the use of FL in email anti phishing. This paper builds upon a deep neural network model, particularly RNN and BERT for phishing email detection. It analyzes the FL entangled learning performance under various settings, including balanced and asymmetrical data distribution. Our results corroborate comparable performance statistics of FL in phishing email detection to centralized learning for balanced datasets, and low organization counts. Moreover, we observe a variation in performance when increasing organizational counts. For a fixed total email dataset, the global RNN based model suffers by a 1.8 accuracy drop when increasing organizational counts from 2 to 10. In contrast, BERT accuracy rises by 0.6 when going from 2 to 5 organizations. However, if we allow increasing the overall email dataset with the introduction of new organizations in the FL framework, the organizational level performance is improved by achieving a faster convergence speed. Besides, FL suffers in its overall global model performance due to highly unstable outputs if the email dataset distribution is highly asymmetric.
[ 2, 14 ]
cs.HC cs.AI
Talking with Robots Opportunities and Challenges. Notwithstanding the tremendous progress that is taking place in spoken language technology, effective speech based human robot interaction still raises a number of important challenges. Not only do the fields of robotics and spoken language technology present their own special problems, but their combination raises an additional set of issues. In particular, there is a large gap between the formulaic speech that typifies contemporary spoken dialogue systems and the flexible nature of human human conversation. It is pointed out that grounded and situated speech based human robot interaction may lead to deeper insights into the pragmatics of language usage, thereby overcoming the current habitability gap'.
[ 32, 68 ]
cs.DM cs.DS
Parameterized algorithms for Partial vertex covers in bipartite graphs. In the weighted partial vertex cover problem (WPVC), we are given a graph G (V,E) , cost function c V rightarrow N , profit function p E rightarrow N , and positive integers R and L . The goal is to check whether there is a subset V' subseteq V of cost at most R , such that the total profit of edges covered by V' is at least L . In this paper we study the fixed parameter tractability of WPVC in bipartite graphs (WPVCB). By extending the methods of Amini et al., we show that WPVCB is FPT with respect to R if c equiv 1 . On the negative side, it is W 1 hard for arbitrary c , even when p equiv 1 . In particular, WPVCB is W 1 hard parameterized by R . We complement this negative result by proving that for bounded degree graphs WPVC is FPT with respect to R . The same result holds for the case of WPVCB when we allow to take only one fractional vertex. Additionally, we show that WPVC is FPT with respect to L . Finally, we discuss a variant of PVCB in which the edges covered are constrained to include a matching of prescribed size and derive a paramterized algorithm for the same.
[ 11, 6 ]
physics.flu-dyn physics.app-ph
Entrainment flow of emerging jet in a half space with the no slip boundary condition. Current development of micro scale technologies increases the interest to viscous flows with low and moderate Reynolds numbers. This work theoretically studies the entrainment flow of a viscous jet emerging from a plane wall into a half space. Methods are developed to find physically meaningful similarity solutions of the Navier Stokes equations with finite values of mass and momentum fluxes and having no characteristic size. Two models are considered flow dominated by momentum flux and flow dominated by mass flux. A new one parameter set of momentum dominated similarity solutions is found. Two linearly independent mass dominated similarity solutions are found. Algorithms are proposed to evaluate the parameters of the similarity models from the mass and momentum balances. Distributions of flow parameters and stresses at the wall are calculated for the similarity models. They are compared with the corresponding distributions obtained by computational fluid dynamics for a more realistic model with a finite size of the jet origin and competitive influence of the mass and momentum fluxes. Such comparison validates the mass dominated similarity model at the jet Reynolds number Re 10 and the momentum dominated similarity model at Re 30. The obtained results are applied to the problem of evaporation at selective laser melting. It is shown that the theoretically estimated flow velocity corresponds to the experimentally observed one. The theory explains formation of the experimentally observed denuded zone and its widening with decreasing the pressure of the ambient gas.
[ 77 ]
math.LO
Congruence pairs of principal MS algebras and perfect extensions. The notion of a congruence pair for principal MS algebras, simpler than the one given by Beazer for K 2 algebras cite 6 , is introduced. It is proved that the congruences of the principal MS algebras L correspond to the MS congruence pairs on simpler substructures L circ circ and D(L) of L that were associated to L in cite 4 . An analogy of a well known Gr "atzer's problem cite Problem 57 11 formulated for distributive p algebras, which asks for a characterization of the congruence lattices in terms of the congruence pairs, is presented here for the principal MS algebras (Problem 1). Unlike a recent solution to such a problem for the principal p algebras in cite 2 , it is demonstrated here on the class of principal MS algebras, that a possible solution to the problem, though not very descriptive, can be simple and elegant. As a step to a more descriptive solution of Problem 1, a special case is then considered when a principal MS algebra L is a perfect extension of its greatest Stone subalgebra L S . It is shown that this is exactly when de Morgan subalgebra L circ circ of L is a perfect extension of the Boolean algebra B(L) . Two examples illustrating when this special case happens and when it does not are presented.