Learning with a drifting target concept

DOI10.1007/978-3-319-24486-0_10MaRDI QIDQ2835624zbMATH OpenOpenAlexFDO

Authors Varun Kanade, Liu Yang, Steve Hanneke

Publication date 30 November 2016

Published in Lecture Notes in Computer Science (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1505.05215

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Learning and adaptive systems in artificial intelligence (68T05)

Abstract: We study the problem of learning in the presence of a drifting target concept. Specifically, we provide bounds on the error rate at a given time, given a learner with access to a history of independent samples labeled according to a target concept that can change on each round. One of our main contributions is a refinement of the best previous results for polynomial-time algorithms for the space of linear separators under a uniform distribution. We also provide general results for an algorithm capable of adapting to a variable rate of drift of the target concept. Some of the results also describe an active learning variant of this setting, and provide bounds on the number of queries for the labels of points in the sequence sufficient to obtain the stated bounds on the error rates.

Recommendations

• On the complexity of learning from drifting distributions
• Learning changing concepts by exploiting the structure of change
• New analysis and algorithm for learning with drifting distributions
• Predictive learning models for concept drift
• scientific article; zbMATH DE number 1301796

Cites work

• Active learning via perfect selective classification
• Activized Learning: Transforming Passive to Active with Improved Label Complexity
• Agnostic active learning
• Analysis of perceptron-based active learning
• scientific article; zbMATH DE number 1332320 (Why is no real title available?)
• Learning changing concepts by exploiting the structure of change
• Learning with a drifting target concept
• Margin Based Active Learning
• On the complexity of learning from drifting distributions
• On the Uniform Convergence of Relative Frequencies of Events to Their Probabilities
• Predicting \(\{ 0,1\}\)-functions on randomly drawn points
• Teaching Dimension and the Complexity of Active Learning
• The complexity of learning according to two models of a drifting environment
• Tracking drifting concepts by minimizing disagreements

Cited in

(12)

• Statistical mechanics of online learning of drifting concepts: A variational approach
• On the complexity of learning from drifting distributions
• Learning changing concepts by exploiting the structure of change
• The complexity of learning according to two models of a drifting environment
• Learning with a drifting target concept
• New analysis and algorithm for learning with drifting distributions
• Learning in environments with unknown dynamics: towards more robust concept learners
• scientific article; zbMATH DE number 1301796 (Why is no real title available?)
• Learning drifting concepts with neural networks
• Advances in Artificial Intelligence – SBIA 2004
• Predictive learning models for concept drift
• Finite sample learning of moving targets

This page was built for publication: Learning with a drifting target concept

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2835624)