Lecture: Aja Huang (Google DeepMind) on AlphaGo
How did Google’s computer programme AlphaGo become so good at board game Go that it could defeat the world champion? On June 29, developer Aja Huang will speak about this during a lecture in the Gorlaeus building.
Aja Huang, Senior Research Scientist at Google DeepMind, is one of the lead developers of AlphaGo. This computer programme surprised the world earlier this year by defeating the world champion in the extremely complicated oriental board game Go. The victory is regarded an important step in the development of Artificial Intelligence.
New search algorithm
During his lecture on June 29, Huang will speak about how AlphaGo has become so powerful. By making use of different types of deep neural networks (deep learning), the programme is able to evaluate the positions on the board and to select new moves. These networks were trained by a novel combination of methods, in which the computer learned in different ways: by analyzing games from human Go players and by self-play.
Furthermore, Huang will talk about a newly developed search algorithm, which was used by AlphaGo to defeat other Go computer programmes, as well as world champion Lee Sedol.
Computers and Games
Aja Huang’s lecture is part of the annual Computers and Games Conference, which is held at the Leiden Institute of Advanced Computer Science (LIACS) this year. During the conference, researchers from around the world will present new results of their research on computer games. At the same time, related events will take place, such as the World Computer Chess Championship and the Computer Olympiad, in which participants let their own computer programmes compete in games such as backgammon, checkers and diplomacy.
(JvdB)
Aja Huang’s keynote lecture is open to everyone and will take place on June 29 from 14:00 to 15:00 in hall 1 of the Gorlaeus building. Registration is not necessary, but due to the limited seating capacity of the venue it is advisable to come early.