Reinforcement learning. Implementation and comparison of approaches

Introduction

The goal of the project is to implement and compare 2 algorithms of offline reinforcement learning: Implicit Q-Learning [1] and Conservative Q-Learning [2].

References

1. Offline Reinforcement Learning with Implicit Q-Learning
2. Conservative Q-Learning for Offline Reinforcement Learning