Potential-based algorithms in on-line prediction and game theory

DOI10.1023/A:1022901500417zbMath1026.68152WikidataQ59538603 ScholiaQ59538603MaRDI QIDQ1394787

Nicolò Cesa-Bianchi, Gábor Lugosi

Publication date: 25 June 2003

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

zbMATH Keywords

perceptron algorithmboostingon-line learninguniversal predictionBlackwell's strategyinternal regretweighted average predictors

Mathematics Subject Classification ID

Nonnumerical algorithms (68W05) Game theory (91A99)

Related Items (21)

Learning correlated equilibria in games with compact sets of strategies ⋮ Improved second-order bounds for prediction with expert advice ⋮ Global Nash convergence of Foster and Young's regret testing ⋮ The query complexity of correlated equilibria ⋮ Dynamic benchmark targeting ⋮ A PDE Approach to the Prediction of a Binary Sequence with Advice from Two History‐Dependent Experts ⋮ No-regret algorithms in on-line learning, games and convex optimization ⋮ Arbitrage of forecasting experts ⋮ Playing against no-regret players ⋮ Forecasting electricity consumption by aggregating specialized experts ⋮ Approachability, regret and calibration: implications and equivalences ⋮ Randomized prediction of individual sequences ⋮ Consistency of discrete Bayesian learning ⋮ Exponential weight algorithm in continuous time ⋮ Note on universal conditional consistency ⋮ How long to equilibrium? The communication complexity of uncoupled equilibrium procedures ⋮ Opinion dynamics with limited information ⋮ Internal regret in on-line portfolio selection ⋮ Internal regret in on-line portfolio selection ⋮ Exponential weight approachability, applications to calibration and regret minimization ⋮ Frequency-based ensemble forecasting model for time series forecasting

This page was built for publication: Potential-based algorithms in on-line prediction and game theory