Regret, Equilibrium, and Learning in Games: A Comprehensive Survey
bronzeagepapi · x · 2026-08-12
Panayotis Mertikopoulos published a survey on learning theory in games, bridging machine learning and economics. The paper covers both single-agent sequential decision-making in adversarial environments and multi-agent interactions.
It focuses on regularized learning policies that encourage exploration via penalties. Key contributions include regret bounds for adversarial multi-armed bandits, equilibrium convergence results in zero-sum games, and a 'folk theorem' linking Nash equilibria to the stability of learning dynamics.
More from Research
- MatBrain splits reasoning from tool use: two models screen 30,000 crystal candidates in 48 hours — bravo_abad · 2026-09-23
- Scale AI launches SWE-Bench Pro V2, a harder agentic coding benchmark — bigblueboo · 2026-09-23
- If AI Writes All the Papers, Peer Review Becomes Humanity's Remaining Role — sudoraohacker · 2026-09-23
- Yarin Gal: I Ignore Papers Where the Candidate Isn't First or Last Author — yaringal · 2026-09-23
- New paper: Transferring the Intelligence of VLMs to Robotic Control — _akhaliq · 2026-09-23
- NTU UMM study: generation training boosts understanding in native multimodal models, but naive sharing conflicts — jiqizhixin · 2026-09-23