Exponential inequalities for nonstationary Markov chains
From MaRDI portal
Publication:2178936
DOI10.1515/DEMO-2019-0007zbMATH Open1434.60171arXiv1808.08811OpenAlexW3105839420WikidataQ115514485 ScholiaQ115514485MaRDI QIDQ2178936FDOQ2178936
Authors: Pierre Alquier, Paul Doukhan, Xiequan Fan 
Publication date: 12 May 2020
Published in: Dependence Modeling (Search for Journal in Brave)
Abstract: Exponential inequalities are main tools in machine learning theory. To prove exponential inequalities for non i.i.d random variables allows to extend many learning techniques to these variables. Indeed, much work has been done both on inequalities and learning theory for time series, in the past 15 years. However, for the non independent case, almost all the results concern stationary time series. This excludes many important applications: for example any series with a periodic behavior is non-stationary. In this paper, we extend the basic tools of Dedecker and Fan (2015) to nonstationary Markov chains. As an application, we provide a Bernstein-type inequality, and we deduce risk bounds for the prediction of periodic autoregressive processes with an unknown period.
Full work available at URL: https://arxiv.org/abs/1808.08811
Recommendations
- Exponential concentration inequalities for additive functionals of Markov chains
- New Bernstein and Hoeffding type inequalities for regenerative Markov chains
- New Bernstein's inequalities for dependent observations and applications to learning theory
- Hoeffding's inequality for uniformly ergodic Markov chains
- Generalization bounds for non-stationary mixing processes
model selectiontime series forecastingmartingalesoracle inequalitiesexponential inequalitiesstatistical learning theorynonstationary Markov chains
Markov processes: estimation; hidden Markov models (62M05) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Inference from stochastic processes and prediction (62M20) Inequalities; stochastic orderings (60E15) Computational learning theory (68Q32) Discrete-time Markov processes on general state spaces (60J05)
Cites Work
- Slope heuristics: overview and implementation
- Prediction of time series by statistical learning: general losses and fast rates
- Concentration inequalities. A nonasymptotic theory of independence
- Asymptotic Theory of Weakly Dependent Random Processes
- Title not available (Why is that?)
- A general class of exponential inequalities for martingales and ratios
- On the Kullback-Leibler information divergence of locally stationary processes
- Nonparametric risk bounds for time-series forecasting
- Exponential inequalities for martingales, with application to maximum likelihood estimation for counting processes
- A Bernstein type inequality and moderate deviations for weakly dependent sequences
- Minimal penalties for Gaussian model selection
- Concentration inequalities and model selection. Ecole d'Eté de Probabilités de Saint-Flour XXXIII -- 2003.
- Weak dependence. With examples and applications.
- Bernstein inequality and moderate deviations under strong mixing conditions
- PAC-Bayesian Inequalities for Martingales
- Probability and moment inequalities for sums of weakly dependent random variables, with applications
- A tail inequality for suprema of unbounded empirical processes with applications to Markov chains
- Iterated Random Functions
- Curvature, concentration and error estimates for Markov chain Monte Carlo
- On McDiarmid's concentration inequality
- A Bernstein-type inequality for some mixing processes and dynamical systems with an application to learning
- Concentration inequalities for Markov chains by Marton couplings and spectral methods
- Nonparametric time series prediction through adaptive model selection
- Sharp bounds for the tails of functionals of Markov chains
- Minimum complexity regression estimation with weakly dependent observations
- Learning from dependent observations
- Concentration of measure inequalities for Markov chains and \(\Phi\)-mixing processes.
- Concentration inequalities for dependent random variables via the martingale method
- Laplace transform estimates and deviation inequalities
- Model selection for weakly dependent time series forecasting
- Exponential inequalities for dynamical measures of expanding maps of the interval
- Deviation inequalities for separately Lipschitz functionals of iterated random functions
- Concentration inequalities for sums and martingales
- Extensions of the Hoeffding-Azuma inequalities
- Deviation inequalities for sums of weakly dependent time series
- Optimal model selection for density estimation of stationary data under various mixing condi\-tions
- Rademacher complexity for Markov chains: applications to kernel smoothing and Metropolis-Hastings
- Stochastic models for time series
- Simpler PAC-Bayesian bounds for hostile data
- Non-parametric estimation of time varying AR(1)-processes with local stationarity and periodicity
- Exponential inequalities for unbounded functions of geometrically ergodic Markov chains: applications to quantitative error bounds for regenerative Metropolis algorithms
- New Bernstein and Hoeffding type inequalities for regenerative Markov chains
- Fast learning from \(\alpha\)-mixing observations
- Concentration of weakly dependent Banach-valued sums and applications to statistical learning methods
Cited In (6)
- Concentration inequalities for non-causal random fields
- High-dimensional VAR with low-rank transition
- Concurrent neural network: a model of competition between times series
- Deviation inequalities for stochastic approximation by averaging
- A deviation inequality for non-reversible Markov processes
- Robbins-Monro algorithm with \(\psi\)-mixing random errors
Uses Software
This page was built for publication: Exponential inequalities for nonstationary Markov chains
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2178936)
