Sequential complexities and uniform martingale laws of large numbers

DOI10.1007/s00440-013-0545-5zbMath1356.60029OpenAlexW2029922822WikidataQ114229372 ScholiaQ114229372MaRDI QIDQ2257118

Alexander Rakhlin, Ambuj Tewari, Karthik Sridharan

Publication date: 23 February 2015

Published in: Probability Theory and Related Fields (Search for Journal in Brave)

Full work available at URL: https://repository.upenn.edu/cgi/viewcontent.cgi?article=1066&context=statistics_papers

zbMATH Keywords

empirical processesdependent dataRademacher averagesuniform Glivenko-Cantelli classessequential prediction

Mathematics Subject Classification ID

Inequalities; stochastic orderings (60E15) Martingales with discrete parameter (60G42) Strong limit theorems (60F15) Combinatorial probability (60C05) Martingales with continuous parameter (60G44) Multistage and repeated games (91A20)

Related Items (9)

Generalization bounds for non-stationary mixing processes ⋮ Discrepancy-based theory and algorithms for forecasting non-stationary time series ⋮ Spectral complexity-scaled generalisation bound of complex-valued neural networks ⋮ Empirical risk minimization and complexity of dynamical models ⋮ Asymptotic sequential Rademacher complexity of a finite function class ⋮ The Convex Mixture Distribution: Granger Causality for Categorical Time Series ⋮ Unnamed Item ⋮ Dimension independent excess risk by stochastic gradient descent ⋮ Portfolio selection in non-stationary markets

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• The Glivenko-Cantelli problem, ten years later
• Some limit theorems for empirical processes (with discussion)
• Uniform and universal Glivenko-Cantelli classes
• Combinatorics of random processes and sections of convex bodies
• A new convex objective function for the supervised learning of single-layer neural networks
• Uniform convergence of Vapnik-Chervonenkis classes under ergodic sampling
• Rates of convergence for classes of functions: The non-i.i.d. case
• The Glivenko-Cantelli problem
• Martingales with values in uniformly convex spaces
• Empirical discrepancies and subadditive processes
• Rates of convergence for empirical processes of stationary mixing sequences
• Efficient distribution-free learning of probabilistic concepts
• Entropy and the combinatorial dimension
• A note on uniform laws of averages for dependent processes
• Optimum bounds for the distributions of martingales in Banach spaces
• Weak convergence and empirical processes. With applications to statistics
• Fat-shattering and the learnability of real-valued functions
• The universal Glivenko-Cantelli property
• A combinatorial problem; stability and order for models and theories in infinitary languages
• On the density of families of sets
• Uniform Central Limit Theorems
• Scale-sensitive dimensions, uniform convergence, and learnability
• The Generic Chaining
• 10.1162/153244303321897690
• On the Uniform Convergence of Relative Frequencies of Events to Their Probabilities

This page was built for publication: Sequential complexities and uniform martingale laws of large numbers