All Projects → georgesung → deep_rl_acrobot

georgesung / deep_rl_acrobot

Licence: MIT license
TensorFlow A2C to solve Acrobot, with synchronized parallel environments

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to deep rl acrobot

Deep-Reinforcement-Learning-With-Python
Master classic RL, deep RL, distributional RL, inverse RL, and more using OpenAI Gym and TensorFlow with extensive Math
Stars: ✭ 222 (+593.75%)
Mutual labels:  deep-reinforcement-learning, openai-gym, policy-gradient, a3c, ddpg
Pytorch Rl
This repository contains model-free deep reinforcement learning algorithms implemented in Pytorch
Stars: ✭ 394 (+1131.25%)
Mutual labels:  deep-reinforcement-learning, openai-gym, policy-gradient, ddpg
Btgym
Scalable, event-driven, deep-learning-friendly backtesting library
Stars: ✭ 765 (+2290.63%)
Mutual labels:  deep-reinforcement-learning, openai-gym, policy-gradient, a3c
yarll
Combining deep learning and reinforcement learning.
Stars: ✭ 84 (+162.5%)
Mutual labels:  deep-reinforcement-learning, openai-gym, policy-gradient, a3c
Easy Rl
强化学习中文教程,在线阅读地址:https://datawhalechina.github.io/easy-rl/
Stars: ✭ 3,004 (+9287.5%)
Mutual labels:  deep-reinforcement-learning, policy-gradient, a3c, ddpg
Minimalrl
Implementations of basic RL algorithms with minimal lines of codes! (pytorch based)
Stars: ✭ 2,051 (+6309.38%)
Mutual labels:  deep-reinforcement-learning, a3c, ddpg
Reinforcement learning tutorial with demo
Reinforcement Learning Tutorial with Demo: DP (Policy and Value Iteration), Monte Carlo, TD Learning (SARSA, QLearning), Function Approximation, Policy Gradient, DQN, Imitation, Meta Learning, Papers, Courses, etc..
Stars: ✭ 442 (+1281.25%)
Mutual labels:  deep-reinforcement-learning, policy-gradient, a3c
Rl a3c pytorch
A3C LSTM Atari with Pytorch plus A3G design
Stars: ✭ 482 (+1406.25%)
Mutual labels:  deep-reinforcement-learning, openai-gym, a3c
Hands On Reinforcement Learning With Python
Master Reinforcement and Deep Reinforcement Learning using OpenAI Gym and TensorFlow
Stars: ✭ 640 (+1900%)
Mutual labels:  deep-reinforcement-learning, openai-gym, policy-gradient
Deeprl Tensorflow2
🐋 Simple implementations of various popular Deep Reinforcement Learning algorithms using TensorFlow2
Stars: ✭ 319 (+896.88%)
Mutual labels:  deep-reinforcement-learning, a3c, ddpg
Deep Reinforcement Learning For Automated Stock Trading Ensemble Strategy Icaif 2020
Deep Reinforcement Learning for Automated Stock Trading: An Ensemble Strategy. ICAIF 2020. Please star.
Stars: ✭ 518 (+1518.75%)
Mutual labels:  deep-reinforcement-learning, openai-gym, ddpg
a3c-super-mario-pytorch
Reinforcement Learning for Super Mario Bros using A3C on GPU
Stars: ✭ 35 (+9.38%)
Mutual labels:  deep-reinforcement-learning, openai-gym, a3c
Deep Reinforcement Learning
Repo for the Deep Reinforcement Learning Nanodegree program
Stars: ✭ 4,012 (+12437.5%)
Mutual labels:  deep-reinforcement-learning, openai-gym, ddpg
Mushroom Rl
Python library for Reinforcement Learning.
Stars: ✭ 442 (+1281.25%)
Mutual labels:  deep-reinforcement-learning, openai-gym, ddpg
Lagom
lagom: A PyTorch infrastructure for rapid prototyping of reinforcement learning algorithms.
Stars: ✭ 364 (+1037.5%)
Mutual labels:  deep-reinforcement-learning, policy-gradient, ddpg
Deep Reinforcement Learning With Pytorch
PyTorch implementation of DQN, AC, ACER, A2C, A3C, PG, DDPG, TRPO, PPO, SAC, TD3 and ....
Stars: ✭ 1,345 (+4103.13%)
Mutual labels:  deep-reinforcement-learning, policy-gradient, a3c
Reinforcement Learning
Minimal and Clean Reinforcement Learning Examples
Stars: ✭ 2,863 (+8846.88%)
Mutual labels:  deep-reinforcement-learning, policy-gradient, a3c
Pytorch Ddpg
Implementation of the Deep Deterministic Policy Gradient (DDPG) using PyTorch
Stars: ✭ 272 (+750%)
Mutual labels:  deep-reinforcement-learning, openai-gym, ddpg
Openai lab
An experimentation framework for Reinforcement Learning using OpenAI Gym, Tensorflow, and Keras.
Stars: ✭ 313 (+878.13%)
Mutual labels:  deep-reinforcement-learning, policy-gradient, ddpg
Slm Lab
Modular Deep Reinforcement Learning framework in PyTorch. Companion library of the book "Foundations of Deep Reinforcement Learning".
Stars: ✭ 904 (+2725%)
Mutual labels:  deep-reinforcement-learning, policy-gradient, a3c

Using Deep Reinforcement Learning to Solve Acrobot

NOTE: This project was done a long time ago when I first started with reinforcement learning, so please excuse some conceptual inaccuracies, e.g. I'm not actually using the DDPG algorithm, and I'm not doing the "asynchronous" part of A3C :) Otherwise, I am using policy gradients with actor/critic networks, with advantage (A2C).

This project uses policy gradients with actor/critic networks and parallel environments to solve OpenAI Gym's Acrobot-v1 environment. As of September 20, 2016, the final learned model placed 3rd on the OpenAI Gym Acrobot-v1 leaderboard, with a score of -80.69 ± 1.06 (see "georgesung's algorithm"): https://gym.openai.com/envs/Acrobot-v1

This project is my capstone project for Udacity's Machine Learning Engineer Nanodegree. For the full capstone project report, please see 'Report.pdf'.

Dependencies

The following depenencies are required:

  • Python 2.7/3.5+
  • NumPy
  • Matplotlib
  • OpenAI Gym
  • TensorFlow 0.10.0

How to run

To run the learning agent with pre-set parameter values, run 'python learning_agent.py'. The main reinformcent learning code is located in this file.

To run the parameter search, run 'python search_params.py'. In this file, you can modify the parameter values over which to search.

Once you know your optimal parameters, enter them in 'full_training.py', and run 'python full_training.py'. This will perform the full training process on the model.

To validate your model (make sure results are consistent), run 'python model_eval.py'.

Detailed report

A full detailed report can be found at 'Report.pdf'

Final model

My final trained model is available at 'models/model.ckpt'. This model is saved as a TensorFlow model.

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].