David Silver
CEO of Ineffable Intelligence and UCL professor; formerly led DeepMind's reinforcement learning team
David Silver is the CEO of Ineffable Intelligence and a professor at University College London; Silver's homepage says that until recently Silver led the reinforcement learning team at DeepMind. Silver is first author of the 2017 AlphaZero paper on mastering chess and shogi through self-play reinforcement learning, and a co-author of the 2013 deep Q-network paper on learning to play Atari games from raw pixels. Silver's site also publishes the lectures and Easy21 assignment of a 2015 Advanced Topics course on reinforcement learning (COMPM050/COMPGI13) under a CC-BY-NC 4.0 license.1342
Work
AlphaZero (external site: arxiv.org)Published research
First author of "Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm" (2017), which extends the AlphaGo Zero approach to chess and shogi.3
Playing Atari with Deep Reinforcement Learning (DQN) (external site: arxiv.org)Published research
Co-author of the 2013 paper presenting a convolutional network trained with a variant of Q-learning to learn control policies for Atari 2600 games directly from pixels.4
Advanced Topics 2015: Reinforcement Learning (course) (external site: davidstarsilver.wordpress.com)Published research
Published the lectures, Easy21 assignment and past exam questions of the 2015 Advanced Topics reinforcement learning course (COMPM050/COMPGI13), made available under CC-BY-NC 4.0 except for some third-party content.2
Other documented roles
Roles the sources document but that are not confirmed as current.
- Led the reinforcement learning team, DeepMind (Google DeepMind)1
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.