In Beta

Welcome to my blog. I write about Reinforcement Learning, AI, Python, and more


One of the first algorithms you might encounter when learning about deep reinforcement learning is DQN. As one of the fundamental algorithms in RL, DQN (Deep Q-Network) has paved the way for many advancements in the field. In this blog post I will guide you through the mechanics of the DQN…



In 2017 Deepmind wowed the world when it's algorithm AlphaGo beat the world champion in Go, Ke Jie. This feat was previously thought to be impossible or at least only possible far in the future. Over the next few years Deepmind would continue to improve upon it's initial design. It would add the…


Policy Gradients

Building from the foundation of Q learning we learned in the last article we now explore an on policy algorithm using policy gradients and actor critic methods. DQN relied upon a experience replay memory in order to get it's results. This makes it an Off Policy algorithm. An Off Policy algorithm is…