Joel Veness

Research Scientist, Google DeepMind


I am a 30-something Computer Scientist.

At the moment I am working on various aspects of Monte-Carlo Tree Search, General Atari 2600 Game Playing, universal source coding and reinforcement learning.

PhD Work

I investigated some practical aspects of Marcus Hutter's AIXI agent, a Bayesian optimality notion for reinforcement learning. My thesis is now available.

Curriculum Vitae

My CV is available on request.


I grew up in Penrith, later moving to Sydney to study at UNSW. I then spent 2 years at the University of Alberta as a postdoc under Michael Bowling, and now work for Google DeepMind.

About 2.0

Now using Twitter!