Vera Rubin Reaches Google Cloud Next Month
Google Cloud's Vera Rubin A5X instances arrive next month, with plans for up to 960,000 Rubin GPUs and David Silver's Ineffable Intelligence as first customer.
Read the storyDavid Silver
Co-founder, Ineffable Intelligence; Professor, UCL
David Silver studies reinforcement learning and systems that acquire knowledge through experience. At DeepMind, he helped lead the research behind AlphaGo, AlphaGo Zero and AlphaZero, and contributed to MuZero and mathematical reasoning work. He is now a co-founder of Ineffable Intelligence, pursuing experience-driven learning, and a UCL professor. The company’s ambition to build a general superlearner remains a research objective.
Follow David Silver
Explore the coverage Get AI Secret’s free daily briefing, including David Silver and the wider AI industry.
At a glance
David Silver studies reinforcement learning and systems that acquire knowledge through experience. At DeepMind, he helped lead the research behind AlphaGo, AlphaGo Zero and AlphaZero, and contributed to MuZero and mathematical reasoning work. He is now a co-founder of Ineffable Intelligence, pursuing experience-driven learning, and a UCL professor. The company’s ambition to build a general superlearner remains a research objective.
- Organization
- Ineffable Intelligence
- Academic role
- Professor of Computer Science, UCL
- Known for
- AlphaGo; AlphaGo Zero; AlphaZero; reinforcement learning
- Previously
- Google DeepMind
Sources checked . Selected career background and milestones.
Career, in context.
Selected research, leadership and career milestones, newest first.
Ineffable announces Sequoia partnership
Sequoia announces its partnership with Silver’s London research lab and co-leadership of the first financing round, describing experience-driven reinforcement learning as the company’s central research direction.
Source: Sequoia Capital: Ineffable partnership ↗Explains the decision to build Ineffable
In a dated note published by Ineffable Intelligence, Silver sets out his motivation for creating a research organization dedicated to learning new knowledge through interaction and experience.
Source: Ineffable Intelligence: A Note from Dave ↗Helps guide AlphaProof mathematics work
DeepMind reports silver-medal-equivalent results on IMO problems from AlphaProof and AlphaGeometry 2. Its credits identify Silver as an adviser and a coordinator of the overall project.
Source: Google DeepMind: IMO silver-medal-level results ↗Co-authors MuZero’s research announcement
Silver and collaborators describe MuZero, which learns a model useful for planning and masters Go, chess, shogi and Atari without being supplied with the environment’s rules.
Source: Google DeepMind: MuZero ↗Co-presents the full AlphaZero evaluation
Silver and collaborators publish the full evaluation of AlphaZero, a single learning approach that masters chess, shogi and Go from self-play with the games’ rules.
Source: Google DeepMind: AlphaZero evaluation ↗Co-introduces AlphaGo Zero
Silver and Demis Hassabis announce AlphaGo Zero, developed by a larger research team, demonstrating Go learning through self-play without training on human game examples.
Source: Google DeepMind: AlphaGo Zero ↗AlphaGo defeats Lee Sedol
AlphaGo wins its five-game match against Lee Sedol by four games to one. Silver’s first-person research account presents the result within DeepMind’s wider reinforcement-learning program.
Source: David Silver / Google DeepMind: Deep Reinforcement Learning ↗
Show earlier historyHide earlier history
The latest on David Silver
Newest firstRelated reporting
Briefs, articles and relevant sections, newest first. Explore older coverage below.
More coverage
5 min read · Members
The First Post Human Lab
Plus: BCI Factory, Biology Has Its Garage-Computer Moment
