Mar 14th, 2014, 3:00 pm in Huxley 345

How to Coach Robots to Play Soccer

People use different methods to make decisions. “Trial and error”, for instance, is widely used by people to learn the best decisions from experiences. Also, arguing with other people, or even self-arguing, could help people to identify advantages and disadvantages of each choice. In this talk, I will introduce how these two techniques can be used jointly to help computers (autonomous agents) to make decisions.

First I will motivate my research, followed by a high-level description of our integration technique. Also, I will compare the performances of my technique and standard Reinforcement Learning on RoboCup Soccer games, under both single-agent and multi-agent settings. Potential application domains and future work will also be briefly discussed. Videos and concrete examples will be running throughout this talk to instantiate my ideas and techniques.

Gao Yang (Alex) is a third year PhD student in the Computational Logic and Argumentation research group, Imperial College London, under the supervision of Prof. Francesca Toni. His research focuses on integratingdomain knowledge into machine learning so as to improve the learning performance. In particular, he has been working on incorporating Argumentation-based knowledge into Reinforcement Learning. He obtained his BEng degree from Northwestern Polytechnical University, China, where, together with other students, he designed strategies for robot soccer games and their team won the championship in several national and international robot soccer competitions. More about me→