Average-case active learning with costs

DOI10.1007/978-3-642-04414-4_15MaRDI QIDQ3648749zbMATH OpenOpenAlexFDO

Authors Andrew Guillory, Jeff A. Bilmes

Publication date 1 December 2009

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

Full work available at URL https://arxiv.org/abs/0905.2997

Mathematics Subject Classification ID

Computational learning theory (68Q32)

Abstract: We analyze the expected cost of a greedy active learning algorithm. Our analysis extends previous work to a more general setting in which different queries have different costs. Moreover, queries may have more than two possible responses and the distribution over hypotheses may be non uniform. Specific applications include active learning with label costs, active learning for multiclass and partial label queries, and batch mode active learning. We also discuss an approximate version of interest when there are very many queries.

Recommendations

• Active learning for cost-sensitive classification
• Bayesian active learning using arbitrary binary valued queries
• Learning cost-sensitive active classifiers
• Two faces of active learning
• Optimised probabilistic active learning (OPAL). For fast, non-myopic, cost sensitive active classification

Cites work

• Approximating Decision Trees with Multiway Branches
• Approximating Optimal Binary Decision Trees
• Decision trees for entity identification, approximation algorithms and hardness results
• Elements of Information Theory
• scientific article; zbMATH DE number 2185610 (Why is no real title available?)
• Teaching Dimension and the Complexity of Active Learning

Cited in

(17)

• Decision trees for function evaluation: simultaneous optimization of worst and expected cost
• Trading off worst and expected cost in decision tree problems
• Cost-sensitive active learning with a label uniform distribution model
• Cost-sensitive active learning through statistical methods
• Approximating decision trees with value dependent testing costs
• Decision tree classification with bounded number of errors
• Annotation cost-sensitive active learning by tree sampling
• Two faces of active learning
• Learning cost-sensitive active classifiers
• Distributed submodular maximization
• Bayesian active learning using arbitrary binary valued queries
• Adaptive submodular ranking and routing
• Active learning for cost-sensitive classification
• A simple approximation algorithm for optimal decision tree
• Decision trees with short explainable rules
• A finite-horizon approach to active level set estimation
• Optimised probabilistic active learning (OPAL). For fast, non-myopic, cost sensitive active classification

This page was built for publication: Average-case active learning with costs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3648749)