Bayesian Pool-based Active Learning With Abstention Feedbacks
This addresses the problem of efficient learning with uncertain labeling in applications like crowdsourcing, though it is incremental as it builds on existing active learning methods.
The paper tackles pool-based active learning with abstention feedbacks, where labelers can abstain from labeling queries, by developing two new Bayesian greedy algorithms that simultaneously learn classification and abstention rates, achieving near-optimal guarantees with a (1-1/e) constant factor approximation in experiments.
We study pool-based active learning with abstention feedbacks, where a labeler can abstain from labeling a queried example with some unknown abstention rate. This is an important problem with many useful applications. We take a Bayesian approach to the problem and develop two new greedy algorithms that learn both the classification problem and the unknown abstention rate at the same time. These are achieved by simply incorporating the estimated abstention rate into the greedy criteria. We prove that both of our algorithms have near-optimality guarantees: they respectively achieve a ${(1-\frac{1}{e})}$ constant factor approximation of the optimal expected or worst-case value of a useful utility function. Our experiments show the algorithms perform well in various practical scenarios.