Consistency of support vector machines for forecasting the evolution of an unknown ergodic dynamical system from observations with unknown noise

DOI10.1214/07-AOS562zbMath1162.62089arXiv0707.0322WikidataQ59196394 ScholiaQ59196394MaRDI QIDQ1020982

Ingo Steinwart, Marian Anghel

Publication date: 4 June 2009

Published in: The Annals of Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/0707.0322

zbMATH Keywords

consistencysupport vector machinesforecasting dynamical systemsobservational noise model

Mathematics Subject Classification ID

Inference from stochastic processes and prediction (62M20) Computational learning theory (68Q32) Learning and adaptive systems in artificial intelligence (68T05) Special processes (60K99) Time series analysis of dynamical systems (37M10) Nonuniformly hyperbolic systems (Lyapunov exponents, Pesin theory, etc.) (37D25) Neural nets and related approaches to inference from stochastic processes (62M45) Smooth dynamical systems: general theory (37C99)

Related Items

On the need for structure modelling in sequence prediction, Prediction of dynamical time series using kernel based regression and smooth splines, Estimating beta-mixing coefficients via histograms, Fast learning from \(\alpha\)-mixing observations, Empirical risk minimization and complexity of dynamical models, On qualitative robustness of support vector machines, Unnamed Item, Measuring the Capacity of Sets of Functions in the Analysis of ERM, Consistency of maximum likelihood estimation for some dynamical systems, Gibbs posterior convergence and the thermodynamic formalism

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• The thermodynamic formalism for expanding maps
• Noise reduction: Finding the simplest dynamical system consistent with the data
• Learning from dependent observations
• Basic properties of strong mixing conditions. A survey and some open questions
• A noise reduction method for signals from nonlinear systems
• Nonparametric statistics for stochastic processes. Estimation and prediction.
• Noise reduction schemes for chaotic time series
• Nonparametric time series prediction through adaptive model selection
• Limits to classification and regression estimation from ergodic processes
• A distribution-free theory of nonparametric regression
• Exponential inequalities for dynamical measures of expanding maps of the interval
• Nonlinear time series. Nonparametric and parametric methods
• Support-vector networks
• Beneath the noise, chaos
• Denoising deterministic time series
• Consistency and robustness of kernel-based regression in convex risk minimization
• 10.1162/153244302760185252
• An Explicit Description of the Reproducing Kernel Hilbert Spaces of Gaussian RBF Kernels
• Memory-universal prediction of stationary random processes
• Almost sure rates of mixing for i.i.d. unimodal maps
• Thermodynamic Formalism
• Function Classes That Approximate the Bayes Risk
• Convexity, Classification, and Risk Bounds
• Theory of Reproducing Kernels
• Equilibrium states and the ergodic theory of Anosov diffeomorphisms