On-going modifications on this repo (by Dabana)

This repo was forked from the DeepRL-Agents repo by Arthur Juliani. I am currently adapting some of the code to train a neural agent to play Vizdoom.

Most of the work so far has been put into the DRQN-VizDoom Jupyther notebook (and the helper2.py file). This is a double dueling DQN agent for which a recurrent layer is added. I am working on improving this agent by implenting prioritized replay.

I am also working on:

1. Being able to save the models and experience buffers for further training
2. Training an agent to play VizDoom with other input buffers from the game such as labels and depth buffers

Future improvements I will be working on soon are:

1. Optimizing prioritized replay using heap queues for the experience replay buffer
2. Implementing Hindsight Experience Replay
3. Adapting the code GA3C from NVLabs to train an A3C agent on GPU and CPU

Deep Reinforcement Learning Agents (by Juliani)

This repository contains a collection of reinforcement learning algorithms written in Tensorflow. The ipython notebook here were written to go along with a still-underway tutorial series I have been publishing on Medium. If you are new to reinforcement learning, I recommend reading the accompanying post for each algorithm.

The repository currently contains the following algorithms:

• Q-Table - An implementation of Q-learning using tables to solve a stochastic environment problem.
• Q-Network - A neural network implementation of Q-Learning to solve the same environment as in Q-Table.
• Simple-Policy - An implementation of policy gradient method for stateless environments such as n-armed bandit problems.
• Contextual-Policy - An implementation of policy gradient method for stateful environments such as contextual bandit problems.
• Policy-Network - An implementation of a neural network policy-gradient agent that solves full RL problems with states and delayed rewards, and two opposite actions (ie. CartPole or Pong).
• Vanilla-Policy - An implementation of a neural network vanilla-policy-gradient agent that solves full RL problems with states, delayed rewards, and an arbitrary number of actions.
• Model-Network - An addition to the Policy-Network algorithm which includes a separate network which models the environment dynamics.
• Double-Dueling-DQN - An implementation of a Deep-Q Network with the Double DQN and Dueling DQN additions to improve stability and performance.
• Deep-Recurrent-Q-Network - An implementation of a Deep Recurrent Q-Network which can solve reinforcement learning problems involving partial observability.
• Q-Exploration - An implementation of DQN containing multiple action-selection strategies for exploration. Strategies include: greedy, random, e-greedy, Boltzmann, and Bayesian Dropout.
• A3C-Doom - An implementation of Asynchronous Advantage Actor-Critic (A3C) algorithm. It utilizes multiple agents to collectively improve a policy. This implementation can solve RL problems in 3D environments such as VizDoom challenges.