Deep_Q_learning

Learning to play games using just the visual input.

Implemented it from understanding of research paper and network is in my own library. https://github.com/ShivamShrirao/dnn_from_scratch .

Using OpenAI gym as game environment.

Results after overnight training on Colab.

Breakout	Pong

The agent has learned the mechanics of the game, formed a few strategies and is able to consistently score good points.

Actual Reward v/s Q Prediction for breakout

Got inspiration from watching this video https://youtu.be/rFwQDDbYTm4 and read the paper (https://arxiv.org/pdf/1312.5602.pdf).