Simulating your robot brain with a game engine

The future of machine learning is big. Really big. ML will be an important part of the software world for decades to come, in everything from robots, to health, to logistics, to the small and personal world. Machine Learning touches the real world all the time. This session explores how you can simulate enough of the real world with a game engine to do ML. Join us.

Specifically, we will explore:

* how game engines like Unity are on the forefront of ML by allowing us to simulate “just enough” of the real world to be useful
* how sophisticated ML frameworks like TensorFlow can be connected to a game engine to generate useful training data, train agents, and create robot brains which can then be transposed to the real world
* how ML requires so much data and infrastructure that it makes business and engineering sense to simulate things instead of spending money and time before you know if it works

We’ll use the Unity game engine and the popular open source combination of the Unity ML-Agents Toolkit and TensorFlow as a case study to show how game engines can be used to perform agent-based deep reinforcement learning, behavioural cloning, and more, as well as generate simulated data for machine learning.

This is the very cutting edge of the cutting edge! Don’t be left behind.