Final project in ELE680 Deep Neural Networks at UiS, University of Stavanger. Comparison of DQN and Double-DQN.
-
Updated
Jan 4, 2023 - Python
Final project in ELE680 Deep Neural Networks at UiS, University of Stavanger. Comparison of DQN and Double-DQN.
Small minimum size gym environments for testing (and maybe finding edge cases)
Deep Reinforcement Learning agent playing Space Invaders
Library for running multiple experiments on configurable custom environments using RLpyt
C++ implementation of the continuous LunarLander environment.
OpenAI Gym environment for the classic Nintendo game Duck Hunt.
EATED 2018 DQN을 이용한 고전게임 강화학습
Tutorial about implementing custom environments using OpenAI gym framework designed for Software for Intelligent Systems and Artificial Intelligence course.
This fork adds multi-agent social dilemma environments in Gym: CoinGame, (Iterated) Prisonner Dilemma, Stag Hunt, Chicken, Matching Pennies. Gym: A toolkit for developing and comparing reinforcement learning algorithms.
Implementation of Q-Learning, SARSA, and Dyna Q to allow an agent to navigate the FrozenLake-v0 environment from OpenAI.
Developed a Deep Q Network (DQN) for the cartpole balancing problem (a Google gym environment) using screen (pixel) input to allow generalization to other discrete binary problems and expandability into robotics.
Developed TD Actor-Critic and solved Grid-world, Open AI 'Lunar Lander-v2' and 'Cartpole-v1' environments.
Maze environments for reinforcement learning (with OpenAI Gym interface)
Repository for the course project done as part of CS-747 (Foundations of Intelligent & Learning Agents) course at IIT Bombay in Autumn 2022.
AI Models Trained on Game Data
Deep convolutional Q-Learning project powered by Gym
A wrapper for OpenAI gym which enables use of multiple environments in synchrony
Add a description, image, and links to the gym-environments topic page so that developers can more easily learn about it.
To associate your repository with the gym-environments topic, visit your repo's landing page and select "manage topics."