Deep Reinforcement Learning

Implementation of various reinforcement learning algorithms following OpenAI Spinning Up guide.

Algorithms

• Vanilla Policy Gradient (VPG)
• Proximal Policy Optimization (PPO)
• Twin Delayed Deep Deterministic Policy Gradient (TD3)

Solved Environments

Here is a list of environments from OpenAI's gym toolkit solved with the above algorithms (link to video of solution in parenthesis).

• CartPole-v0 (VPG, PPO)
• LunarLander-v2 (PPO)
• Pendulum-v1 (TD3)