Online calibrated forecasts: memory efficiency versus universality for learning in games

DOI10.1007/s10994-006-0219-yzbMath1471.91051OpenAlexW2037905382MaRDI QIDQ2384142

Gürdal Arslan, Shie Mannor, Jeff S. Shamma

Publication date: 20 September 2007

Published in: Machine Learning (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/s10994-006-0219-y

zbMATH Keywords

stochastic approximationcalibrationforecastingfictitious playlearning in gamesODE methodprediction of universal sequences

Mathematics Subject Classification ID

Rationality and learning in game theory (91A26)

Related Items

Regret minimization in repeated matrix games with variable stage duration, Deterministic calibration and Nash equilibrium

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Stochastic uncoupled dynamics and Nash equilibrium
• A game of prediction with expert advice
• Asynchronous stochastic approximation and Q-learning
• Calibrated learning and correlated equilibrium
• Stochastic approximation with two time scales
• Convergent multiple-timescales reinforcement learning algorithms in normal form games
• Regret in the on-line decision problem
• Mixed equilibria and dynamical systems arising from fictitious play in perturbed games
• An easier way to calibrate.
• Convergence rate of linear two-time-scale stochastic approximation.
• Self-Calibrating Priors Do Not Exist
• Asymptotic calibration
• COMPLEXITY AND REAL COMPUTATION: A MANIFESTO
• Evolutionary Games and Population Dynamics
• A separation principle for the control of a class of nonlinear systems
• Learning Theory
• Equivalent necessary and sufficient conditions on noise sequences for stochastic approximation algorithms
• The Nonstochastic Multiarmed Bandit Problem
• The O.D.E. Method for Convergence of Stochastic Approximation and Reinforcement Learning
• Dynamic fictitious play, dynamic gradient play, and distributed convergence to Nash equilibria
• Adaptive Heuristics
• Calibration with Many Checking Rules
• A general class of adaptive strategies