Deviation inequalities for stochastic approximation by averaging

DOI10.1016/j.spa.2022.07.002zbMath1502.60055arXiv2102.08685OpenAlexW3130172249WikidataQ114130694 ScholiaQ114130694MaRDI QIDQ2169079

Xiequan Fan, Paul Doukhan, Pierre Alquier

Publication date: 29 August 2022

Published in: Stochastic Processes and their Applications (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2102.08685

zbMATH Keywords

martingalesempirical risk minimizationiterated random functionsdeviation inequalitiesstochastic approximation by averaging

Mathematics Subject Classification ID

Inequalities; stochastic orderings (60E15) Martingales with discrete parameter (60G42) Discrete-time Markov processes on general state spaces (60J05) Large deviations (60F10)

Cites Work

• Limit theory for moderate deviations from a unit root
• Model selection for weakly dependent time series forecasting
• Hoeffding's inequality for supermartingales
• Deviation inequalities for separately Lipschitz functionals of iterated random functions
• A Bernstein type inequality and moderate deviations for weakly dependent sequences
• On McDiarmid's concentration inequality
• Moment inequalities for sums of dependent random variables under projective conditions
• Nonparametric time series prediction through adaptive model selection
• Large deviations for martingales.
• A general class of exponential inequalities for martingales and ratios
• Exponential inequalities for nonstationary Markov chains
• High-dimensional VAR with low-rank transition
• Some probabilistic inequalities for martingales
• Deviation inequalities for separately Lipschitz functionals of composition of random functions
• Deviation inequalities for martingales with applications
• Fast learning from \(\alpha\)-mixing observations
• Best possible bounds of the von Bahr-Esseen type
• Towards a unified asymptotic theory for autoregression
• Acceleration of Stochastic Approximation by Averaging
• Iterated Random Functions
• Regression Theory for Near-Integrated Time Series
• Large and moderate deviations for the left random walk on GL d (R)
• Prediction of time series by statistical learning: general losses and fast rates
• Large deviations of sums of independent random variables
• Estimates for the distribution of sums and maxima of sums of random variables without the Cramér condition