Turing Lecture: The Role of Multi-Agent Learning in Artificial Intelligence Research at DeepMind

The Role of Multi-agent Learning in Artificial Intelligence Research at DeepMind
Thore Graepel
In computer science, an agent can be thought of as a computational entity that repeatedly perceives the environment, and takes action so as to optimize long term reward. 
We consider intelligence to be the ability of an agent to achieve goals in a wide range of environments (Legg & Hutter). Thinking in evolutionary/ecological terms, the richest environments for a given agent are themselves evolving collections of agents. These could be biological organisms, or companies within a given market.
In this lecture, Thore will discuss the important role multi-agent learning has to play in artificial intelligence research and the challenges it presents.
Specifically, he  will discuss two example projects from multi-agent learning work at DeepMind.
Firstly, Thore will show how to use advances in deep reinforcement learning to study the age-old question of how cooperation arises among self-interested agents. By defining Sequential Social Dilemmas, this work goes beyond simple matrix games such as the famous game theory example of the Prisoner's Dilemma, and can model new aspects of social dilemmas such as temporal dynamics and coordination problems.
Secondly, Thore will discuss the AlphaGo project, in which DeepMind used the multi-agent algorithm of Learning from Self-Play to create the first computer program to beat a top professional Go player at the full-size game of Go, a feat thought to be at least a decade away by Go and AI experts alike.

Biography
Thore GraepelResearch Group Lead at DeepMind and Chair of Machine learning, University College London
Thore Graepel is a research group lead at Google DeepMind and holds a part-time position as Chair of Machine Learning at University College London. He studied physics at the University of Hamburg, Imperial College London, and Technical University of Berlin, where he also obtained his PhD in machine learning in 2001. He spent time as a postdoctoral researcher at ETH Zurich and Royal Holloway College, University of London, before joining Microsoft Research in Cambridge in 2003, where he co-founded the Online Services and Advertising group.
Major applications of Thore’s work include Xbox Live’s TrueSkill system for ranking and matchmaking, the AdPredictor framework for click-through rate prediction in Bing, and the Matchbox recommender system which inspired the recommendation engine of Xbox Live Marketplace. More recently, Thore’s work on the predictability of private attributes from digital records of human behaviour has been the subject of intense discussion among privacy experts and the general public. Thore’s research interests are in artificial intelligence and machine learning and include probabilistic graphical models, reinforcement learning, game theory, and multi-agent systems. He has published over one hundred peer-reviewed papers, is a named co-inventor on dozens of patents, serves on the editorial boards of JMLR and MLJ, and is a founding editor of the book series Machine Learning & Pattern Recognition at Chapman & Hall/CRC. At DeepMind, Thore has returned to his original passion of understanding and creating intelligence, and recently contributed to creating AlphaGo, the first computer program to defeat a human professional player in the full-sized game of Go, a feat previously thought to be at least a decade away.

NOTE: This event has met a high level of interest. Please register your interest to apply for a ticket and we will confirm your space shortly.