Richard Sutton

Richard Sutton appears in E226｜聊聊DeepMind创始人哈萨比斯：一个科学家与失控的AI竞赛 as the reinforcement-learning theorist whose school influenced David Silver and the early DeepMind route. The episode contrasts Sutton’s reinforcement-learning lineage with Geoff Hinton’s deep-learning lineage before describing how DeepMind later fused the two in practice.

In this wiki, Sutton anchors the older agent-and-reward view of AGI. The source uses that lineage to explain why Reinforcement Learning AGI Path was compelling before ChatGPT and Language Model Scaling Bet shifted frontier-model strategy.

Connections

• David Silver and DeepMind — source path from Sutton-style RL into DeepMind.
• Geoff Hinton — deep-learning contrast in the episode.
• Reinforcement Learning AGI Path, AlphaGo, and Agent RL — technical concepts and system examples.
• Language Model Scaling Bet — later route contrast.