Improved second-order bounds for prediction with expert advice

DOI10.1007/s10994-006-5001-7zbMath1471.91074OpenAlexW2005951230WikidataQ59538580 ScholiaQ59538580MaRDI QIDQ2384131

Nicolò Cesa-Bianchi, Yishay Mansour, Gilles Stoltz

Publication date: 20 September 2007

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

Full work available at URL: https://doi.org/10.1007/s10994-006-5001-7

zbMATH Keywords

prediction with expert adviceindividual sequencesexponentially weighted averages

Mathematics Subject Classification ID

Algorithmic game theory and complexity (91A68) Games with incomplete information, Bayesian games (91A27)

Related Items (14)

Adaptive and optimal online linear regression on \(\ell^1\)-balls ⋮ Combining initial segments of lists ⋮ Relaxing the i.i.d. assumption: adaptively minimax optimal regret via root-entropic regularization ⋮ Forecasting electricity consumption by aggregating specialized experts ⋮ Online multiple kernel classification ⋮ Stochastic online convex optimization. Application to probabilistic time series forecasting ⋮ Extracting certainty from uncertainty: regret bounded by variation in costs ⋮ Sequential model aggregation for production forecasting ⋮ Regret to the best vs. regret to the average ⋮ Structural Online Learning ⋮ The Follow Perturbed Leader Algorithm Protected from Unbounded One-Step Losses ⋮ A general equilibrium model of investor sentiment ⋮ AN ONLINE PORTFOLIO SELECTION ALGORITHM WITH REGRET LOGARITHMIC IN PRICE VARIATION ⋮ Exponential weight approachability, applications to calibration and regret minimization

Uses Software

• AdaBoost.MH

Cites Work

• Unnamed Item
• On tail probabilities for martingales
• A game of prediction with expert advice
• The weighted majority algorithm
• A decision-theoretic generalization of on-line learning and an application to boosting
• Potential-based algorithms in on-line prediction and game theory
• Adaptive and self-confident on-line learning algorithms
• Minimizing Regret With Label Efficient Prediction
• How to use expert advice
• The Nonstochastic Multiarmed Bandit Problem
• On‐Line Portfolio Selection Using Multiplicative Updates
• Regret Minimization Under Partial Monitoring
• Algorithmic Learning Theory
• Prediction, Learning, and Games

This page was built for publication: Improved second-order bounds for prediction with expert advice