Difference between revisions of "Demis Hassabis"

From Chessprogramming wiki
Jump to: navigation, search
Line 53: Line 53:
  
 
'''[[People|Up one level]]'''
 
'''[[People|Up one level]]'''
[[Category:Chess Player]]
+
[[Category:Chess Player|Hassabis]]

Revision as of 12:47, 16 August 2018

Home * People * Demis Hassabis

Demis Hassabis [1]

Demis Hassabis,
a British computer scientist, neuroscientist, artificial intelligence researcher, computer game developer, entrepeneur, and founder and CEO of DeepMind [2]. He was a child prodigy in chess starting with age 4, with an Elo rating of 2300 at the age of 13, is an accomplished shogi and poker player [3], and further won the Mind Sports Olympiad in 1999, aged 23, and four more times in consecutive years [4]. At age 8, he baught his first computer, a ZX Spectrum from a prize money of a chess match, and made his first practical AI experiences in developing chess and Othello programs.

At age 16, Demis Hassabis began his game developer career at Bullfrog Productions, working with Peter Molyneux on Theme Park, later continuing his collaboration with Molyneux at Lionhead Studios on Black & White. His further games Republic: The Revolution and Evil Genius were already developed under his 1998 founded Elixir Studios . He graduated in computer science from University of Cambridge in 1997 and defended his Ph.D. from University College London in cognitive neuroscience in 2009, followed by postdocs at MIT and Harvard.

DeepMind

After finishing his academic career, Demis Hassabis founded DeepMind in 2010, which was acquired by Google in 2014. In January 2016, DeepMind reported archiving an AI "breakthrough" with their Go playing program AlphaGo by beating European Go champion Fan Hui [5] in October 2015 with a 5 - 0 score [6] [7] [8]. In December 2017, a breakthrough in the domains of chess and Shogi was reported, combining Deep learning with Monte-Carlo Tree Search [9].

Selected Publications

[10] [11]

2009

2015 ...

2016

2017

External Links

DeepMind

Gaming

AlphaGo

References

Up one level