Part 2 — Building a deep Q-network to play Gridworld — Catastrophic Forgetting and Experience…

Part 2 — Building a deep Q-network to play Gridworld — Catastrophic Forgetting and Experience Replay

In this article let’s talk about the problem in Vanilla Q-learning model: Catastrophic forgetting . We will solve this problem using Experience replay and see the improvement we have made in playing GridWorld

Welcome to the second part of Deep Q-network tutorials. This is the continuation of the part 1. If you have not read the part 1, I strongly suggest to go …

building data science deep learning deep-q-learning machine learning network part reinforcement learning