Weighted last-step min-max algorithm with improved sub-logarithmic regret

DOI10.1016/j.tcs.2014.09.028zbMath1360.68701OpenAlexW2179980973MaRDI QIDQ465257

Edward Moroshko, Koby Crammer

Publication date: 31 October 2014

Published in: Theoretical Computer Science (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.tcs.2014.09.028

zbMATH Keywords

online learningregressionmin-max learning

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Online algorithms; streaming algorithms (68W27)

Uses Software

• AdaGrad

Cites Work

• Adaptive regularization of weight vectors
• Exponentiated gradient versus gradient descent for linear predictors
• An identity for kernel ridge regression
• Prediction in the worst case
• 10.1162/153244301753683726
• Re-adapting the Regularization of Weights for Non-stationary Regression
• Online Regression Competitive with Changing Predictors
• Performance of adaptive estimation algorithms in dependent random environments
• A game theory approach to constrained minimax state estimation
• Competitive On-line Statistics
• A Second-Order Perceptron Algorithm
• Prediction, Learning, and Games
• Relative loss bounds for on-line density estimation with the exponential family of distributions
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item

This page was built for publication: Weighted last-step min-max algorithm with improved sub-logarithmic regret