Implementation of RL Algorithms

Picture from Reinforcement Learning - An Introduction

Abstract

To broaden my perspective on Machine Learning, I took up Reinforcement Learning courses by David Silver, Stanford CS234. To understand the nuances in the field, I implemented basic algorithms. The inspiration for the same was also drawn through a project where the idea of using Imitation Learning to overcome several limitations, struck me.

Implementated:

  • DQN
  • Vanilla Policy Gradient
  • PPO
  • DDPG

Results

Value and Policy Iteration

Method Deterministic Frozen Lake Stochastic Frozen Lake
Value Iteration 7 8
Policy Iteration 7 3

Q-Learning

Deep Q-Learning

Avatar
Khush Agrawal
Senior Year Undergraduate Student

Reinforcement Learning

Related