Plug-in approach to active learning

MaRDI QIDQ5405110zbMATH OpenFDO

Authors Stanislav Minsker

Publication date 1 April 2014

Full work available at URL https://arxiv.org/abs/1104.1450, http://www.jmlr.org/papers/v13/minsker12a.html

zbMATH Keywords

active learningclassificationconfidence bandsmodel selectionselective sampling

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Nonparametric regression and quantile regression (62G08) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Learning and adaptive systems in artificial intelligence (68T05)

Abstract: We present a new active learning algorithm based on nonparametric estimators of the regression function. Our investigation provides probabilistic bounds for the rates of convergence of the generalization error achievable by proposed method over a broad class of underlying distributions. We also prove minimax lower bounds which show that the obtained rates are almost tight.

Recommendations

• Rates of convergence in active learning
• Minimax Bounds for Active Learning
• Toward optimal probabilistic active learning using a Bayesian approach
• Statistical asymptotic theory of active learning
• Probability bounds for active learning in the regression problem

Cited in

(9)

• Statistical asymptotic theory of active learning
• Improving importance estimation in pool-based batch active learning for approximate linear regression
• Surrogate losses in passive and active learning
• Probability bounds for active learning in the regression problem
• Learning from non-iid data: fast rates for the one-vs-all multiclass plug-in classifiers
• Active nearest-neighbor learning in metric spaces
• The relationship between agnostic selective classification, active learning and the disagreement coefficient
• Bayesian active learning using arbitrary binary valued queries
• Active learning for cost-sensitive classification

This page was built for publication: Plug-in approach to active learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5405110)