# Bayesian Active Learning With Abstention Feedbacks

## TL;DR

This paper introduces Bayesian active learning algorithms that simultaneously learn a classification task and an unknown abstention rate, with proven near-optimal guarantees and effective performance in practical scenarios.

## Contribution

Develops two greedy algorithms for Bayesian active learning with abstention feedbacks, incorporating abstention rate estimation and providing near-optimality guarantees.

## Key findings

- Algorithms achieve near-optimal approximation guarantees.
- Perform well in various practical scenarios.
- Effectively learn classification and abstention rate simultaneously.

## Abstract

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 average abstention rate into the greedy criteria. We prove that both 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.