UWEE Tech Report Series

Average-Case Active Learning with Costs


Andrew Guillory, Jeff Bilmes

machine learning, active learning, learning theory


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.

Download the PDF version

Download the Gzipped Postscript version