Learning whenever learning is possible: universal learning under general stochastic processes

MaRDI QIDQ5159394zbMATH OpenFDO

Authors Steve Hanneke

Publication date 27 October 2021

Full work available at URL https://arxiv.org/abs/1706.01418, https://jmlr.csail.mit.edu/papers/v22/17-298.html

zbMATH Keywords

generalizationnon-stationary processesnonparametric estimationonline learningstatistical learning theorystochastic processesuniversal consistency

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Recommendations

• Universal regression with adversarial responses
• Learnability, stability and uniform convergence
• On universal learning algorithms
• Learning theory estimates with observations from general stationary stochastic processes
• ONLINE LEARNING WITH MARKOV SAMPLING

Cites work

• A distribution-free theory of nonparametric regression
• A simple randomized algorithm for sequential prediction of ergodic time series
• A theory of learning from different domains
• An algebraic characterization of measure algebras
• Asymptotically mean stationary measures
• Data-dependent k/sub n/-NN and kernel estimators consistent for arbitrary processes
• Estimation of dependences based on empirical data. Transl. from the Russian by Samuel Kotz
• Generalization bounds for time series prediction with non-stationary processes
• Guessing the next output of a stationary process
• How to use expert advice
• scientific article; zbMATH DE number 3959831 (Why is no real title available?)
• scientific article; zbMATH DE number 3680123 (Why is no real title available?)
• scientific article; zbMATH DE number 3457210 (Why is no real title available?)
• scientific article; zbMATH DE number 1332320 (Why is no real title available?)
• scientific article; zbMATH DE number 638631 (Why is no real title available?)
• scientific article; zbMATH DE number 722611 (Why is no real title available?)
• scientific article; zbMATH DE number 1166155 (Why is no real title available?)
• scientific article; zbMATH DE number 893887 (Why is no real title available?)
• scientific article; zbMATH DE number 1416652 (Why is no real title available?)
• scientific article; zbMATH DE number 3245885 (Why is no real title available?)
• Learning from dependent observations
• Limits to classification and regression estimation from ergodic processes
• Maharam's problem
• Measure theory. Vol. 3. Measure algebras
• Measure theory. Vol. I and II
• Nonparameteric estimation in mixing sequences of random variables
• Nonparametric inference for ergodic, stationary time series
• On consistency in nonparametric estimation under mixing conditions.
• On density estimation from ergodic processes
• On optimal sequential prediction for general processes
• Online learning via sequential complexities
• Pattern recognition for conditionally independent data
• Predicting \(\{ 0,1\}\)-functions on randomly drawn points
• Prediction of random sequences and universal coding
• Prediction, Learning, and Games
• Probability, Random Processes, and Ergodic Properties
• Propri�t�s de convergence presque compl�te du pr�dicteur � noyau
• Rates of convergence for empirical processes of stationary mixing sequences
• Rates of uniform convergence of empirical means with mixing processes
• Regression Estimation from an Individual Stable Sequence
• Sample Selection Bias Correction Theory
• The generalization performance of ERM algorithm with strongly mixing observations
• The strong law of large numbers for sequential decisions under uncertainty
• The universal Glivenko-Cantelli property
• The weighted majority algorithm
• Theory of capacities
• Theory of statistics
• Uniform approximation of Vapnik-Chervonenkis classes
• Uniform convergence of Vapnik-Chervonenkis classes under ergodic sampling
• Universal forecasting algorithms
• Universal linear prediction by model order weighting
• Universal schemes for learning the best nonlinear predictor given the infinite past and side information
• Universal schemes for prediction, gambling and portfolio selection

Cited in

(9)

• Universal Bayes consistency in metric spaces
• scientific article; zbMATH DE number 3874345 (Why is no real title available?)
• Universal regression with adversarial responses
• Relaxing the i.i.d. assumption: adaptively minimax optimal regret via root-entropic regularization
• Learning Whenever Learning is Possible: Universal Learning under General Stochastic Processes
• Approximation with ergodic processes and testability
• Countable alphabet stationary processes with at least one memory word and intermittent estimation with universal rates
• Universally consistent online learning with arbitrarily dependent responses
• Universal online learning with unbounded losses: memory is all you need

This page was built for publication: Learning whenever learning is possible: universal learning under general stochastic processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5159394)