|
--- |
|
tags: |
|
- CartPole-v1 |
|
- reinforce |
|
- reinforcement-learning |
|
- custom-implementation |
|
- deep-rl-class |
|
model-index: |
|
- name: CartPole-v1 |
|
results: |
|
- task: |
|
type: reinforcement-learning |
|
name: reinforcement-learning |
|
dataset: |
|
name: CartPole-v1 |
|
type: CartPole-v1 |
|
metrics: |
|
- type: mean_reward |
|
value: 500.00 +/- 0.00 |
|
name: mean_reward |
|
verified: false |
|
--- |
|
|
|
# **Reinforce** Agent playing **CartPole-v1** |
|
I have used Reinforcement learning in a game, Cart Pole. The aim is to keep the equilibrium by moving left/right. While training, the game uses its results/rewards to modify its parameters to get more rewards. Specifically, the model learns what kind of tactics let the cart pole balance, and as it fails, it learns and applies those tactics to balance the cart pole. |
|
|
|
Some links I've found helpful include: |
|
https://huggingface.co/learn/deep-rl-course/en/unit0/introduction#certification-process |
|
https://colab.research.google.com/github/huggingface/deep-rl-class/blob/master/notebooks/unit4/unit4.ipynb#scrollTo=NCNvyElRStWG |
|
https://spinningup.openai.com/en/latest/spinningup/rl_intro3.html#don-t-let-the-past-distract-you |
|
https://stable-baselines3.readthedocs.io/en/master/guide/rl_tips.html |
|
https://gymnasium.farama.org/content/migration-guide/ |
|
https://github.com/enerrio/CartPole-Reinforcement-Learning |
|
https://www.ibm.com/topics/overfitting |
|
https://learningds.org/ch/04/modeling_loss_functions.html |
|
https://www.geeksforgeeks.org/reinforce-algorithm/ |
|
|