Part 1 — Building a deep Q-network to play Gridworld — DeepMind’s deep Q-networks

Part 1 — Building a deep Q-network to play Gridworld — DeepMind’s deep Q-networks

In this article let’s build a Deep Q-network similar to the DeepMind’s Atari agent to play Gridworld problem. We will build virtually the same system the DeepMind did from scratch to understand the Deep Q-network in detail. We will understand the drawbacks of vanilla Deep Q-network and come up with the clever ways to overcome them.

We have seen how to build a simple Deep Reinforcement …

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