BeeGass/Deep-Q-Learning

This is my attempt at implementing the paper "Playing Atari with Deep Reinforcement Learning" By Volodymyr Mnih, Koray Kavukcuoglu, David Silver, Alex Graves, Ioannis Antonoglou, Daan Wierstra and Martin Riedmiller. This is my first attempt at both reading and implementing a research paper.

Deep_Q-Learning

What Files Should I Look At

• There has been poor use of version control due in part to the newness of google colab to the group.
• To view code associated with the latest iteration in our approach to DQN please refer to Deep_Q_Network.ipynb. If you want to view the visualization to portions of DQN please refer to Manim_For_DQN.ipynb

Resources That Helped

• Gym Documentation Is Fantastic
• cyoon1729's "vanillaDQN" has some incredibly clear and concise code
• Chris Yoon's, aka cyoon1729, Medium Article On DQN
• Deep Q-Network (DQN)-I
• Deep Q-Network (DQN)-II
• PyTorch's REINFORCEMENT LEARNING (DQN) TUTORIAL
• PyTorch's TRAIN A MARIO-PLAYING RL AGENT
• The Great David Silver And His Lectures On Reinforcement Learning
• Really wonderful resource that shows side by side explanation of code
• Another Great Tutorial With Code
• For When You Want To Learn About Epsilon Decay
• Of Course The Deep Q Network Paper

Thanks

Thank you to

• @edkazcarlson
• @moneill0
• @20hub
• @yichenlilyc
• @dtkatch

for the late night/early morning help

Clips

DQN-TOC.mp4

chess.mp4

Q-table.mp4

Q-Leaning-tic-tac-toe.mp4

Q-table.With.Q.Val.mp4

DQN-tic-tac-toe.mp4

DQN-psuedo-code.mp4

init-replay-buffer.mp4

nn-init-weights.mp4

store-trans.mp4

random-action.mp4

evict-trans.mp4

Q-Val.explain.mp4