Average-Case Active Learning with Costs

2009/05/18 by Andrew Guillory, Guillory, Andrew, Jeff Bilmes +1 · 4 citations
Computer Science · Decision Sciences · #Machine Learning and Algorithms #Optimization and Search Problems #Advanced Bandit Algorithms Research

paper · pdf · doi:10.48550/arxiv.0905.2997

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.

Cited by

• How to Purchase Labels? A Cost-Effective Approach Using Active Learning Markets
• Trading off Worst and Expected Cost in Decision Tree Problems and a\n Value Dependent Model
• Structural query-by-committee
• A Simple Approximation Algorithm for Optimal Decision Tree

Related

• Bounds on the Generalization Error in Active Learning
• Cost-Based Budget Active Learning for Deep Learning
• Active Learning with Importance Sampling
• Improve Cost Efficiency of Active Learning over Noisy Dataset
• Fast Rates in Pool-Based Batch Active Learning