Understanding Neural Networks by Playing Games

The objective of this workshop is to get the participants to understand CNNs by playing the simple Chrome T-Rex game. An intuitive and simple to follow tutorial that requires no prior knowledge.

Tags: Artificial Intelligence, Deep Learning & Artificial Intelligence

Scheduled on thursday 10:30 in room openhub


Sidharth Ramachandran (sidhusmart)

Data Scientist focused on solving real business problems through innovative data products


Playing the Chrome T-Rex involves generating training data by actually playing the game in Chrome offline mode. We need roughly 300-500 images for training and this can be easily generated by playing the game just 5 times. We then work through the basics of Neural Networks and CNNs and then actually training and testing a model. This model can then be deployed easily within the browser and participants can see the model playing the game (with reasonable accuracy) - all within a manner of an hour. We then take this one step further by trying to understand what the model and how it correlates with how they played the game in the first place. This helps to build an understanding of how humans think and what the model actually learns and looks for in the image.