Online learning via sequential complexities

MaRDI QIDQ5502109zbMATH OpenFDO

Authors Alexander Rakhlin, Karthik Sridharan, Ambuj Tewari

Publication date 17 August 2015

Full work available at URL http://jmlr.csail.mit.edu/papers/v16/rakhlin15a.html

zbMATH Keywords

online learningregret minimizationsequential complexities

Mathematics Subject Classification ID

Inference from stochastic processes and prediction (62M20) Learning and adaptive systems in artificial intelligence (68T05) Rationality and learning in game theory (91A26)

Recommendations

• On martingale extensions of Vapnik-Chervonenkis theory with applications to online learning
• Structural online learning
• scientific article; zbMATH DE number 1569102
• Online learning and game theory. A quick overview with recent results and applications
• scientific article; zbMATH DE number 7564348

Cited in

(26)

• On-line learning and the metrical task system problem
• Dimension independent excess risk by stochastic gradient descent
• Discrepancy-based theory and algorithms for forecasting non-stationary time series
• Exponential weight approachability, applications to calibration and regret minimization
• Tractability of batch to sequential conversion
• On martingale extensions of Vapnik-Chervonenkis theory with applications to online learning
• Structural online learning
• The complexity of learning SUBSEQ(A)
• On-line learning in parity machines
• scientific article; zbMATH DE number 7014216 (Why is no real title available?)
• scientific article; zbMATH DE number 1569102 (Why is no real title available?)
• scientific article; zbMATH DE number 7306919 (Why is no real title available?)
• Learning whenever learning is possible: universal learning under general stochastic processes
• Asymptotic sequential Rademacher complexity of a finite function class
• Opportunistic approachability and generalized no-regret problems
• Online learning of quantum states
• Portfolio selection in non-stationary markets
• Universal regression with adversarial responses
• Model theory and agnostic online learning via excellent sets
• Online infinite-dimensional regression: learning linear operators
• Multiclass online learnability under bandit feedback
• Near-tight closure bounds for the Littlestone and threshold dimensions
• Sequential prediction under log-loss with side information
• Universally consistent online learning with arbitrarily dependent responses
• A characterization of multioutput learnability
• Adversarial laws of large numbers and optimal regret in online classification

This page was built for publication: Online learning via sequential complexities

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5502109)