Opportunistic Approachability and Generalized No-Regret Problems

DOI10.1287/moor.2014.0643zbMath1310.91029MaRDI QIDQ5247611

Andrey Bernstein, Shie Mannor, Nahum Shimkin

Publication date: 24 April 2015

Published in: Mathematics of Operations Research (Search for Journal in Brave)

Full work available at URL: https://semanticscholar.org/paper/964d31995f1f3a9c02255c5df803a2456a45a9f4

zbMATH Keywords

calibrated play; no-regret algorithms; Blackwell's approachability

Mathematics Subject Classification ID

91B06: Decision theory

68T05: Learning and adaptive systems in artificial intelligence

91A05: 2-person games

91A80: Applications of game theory

91A20: Multistage and repeated games

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• A general internal regret-free strategy
• Continuous selections. I
• An analog of the minimax theorem for vector payoffs
• Regret minimization in repeated matrix games with variable stage duration
• Calibrated learning and correlated equilibrium
• Approachability in infinite dimensional spaces
• Calibrated forecasting and merging
• Approachability with bounded memory
• Approachable sets of vector payoffs in stochastic games
• A Geometric Proof of Calibration
• On a Unified Framework for Approachability with Full or Partial Monitoring
• Attainability in Repeated Games with Vector Payoffs
• Uniform conditional variability ordering of probability distributions
• Sensitivity of the Stationary Distribution of a Markov Chain
• Guaranteed performance regions in Markovian systems with competing decision makers
• Prediction, Learning, and Games
• Approachability in a Two-person Game
• A Necessary and Sufficient Condition for Approachability
• The Empirical Bayes Envelope and Regret Minimization in Competitive Markov Decision Processes
• A general class of adaptive strategies