Back

Speaker "Yuan Gao" Details Back

 

Topic

Training Game AI using Deep Reinforcement Learning

Abstract

Many of the recent advances in deep reinforcement learning have stemmed from video games. In this session, we’ll explore a brief history of this relationship. We’ll also show how Unity, the world’s leading game engine, are leveraging cutting edge research to solve gaming’s biggest challenges through the Unity Machine Learning Agents Toolkit (one of the most popular open source toolkits for deep learning). And how we are driving the boundaries of new AI research through the Obstacle Tower Challenge.
Who is this presentation for?
This session is for anyone interested in using deep reinforcement learning to train your own AI in games.
Prerequisite knowledge:

What you'll learn?

Profile

Yuan Gao is a machine learning engineer at Unity Technologies. He is one of the core contributors of Unity Machine Learning Agents Toolkit, a open-source Unity plugin that enables games and simulations to serve as environments for training intelligent agents.One of his recent projects explored the possibility of training agents to play advanced levels of Jam City’s Bubble Shooter game using deep reinforcement learning enabled by the toolkit. Prior to joining Unity, he worked in the Data Science Team in Fanatics. He received his MS degree in Computer Science at the University of Chicago.