Active Learning from Weak and Strong Labelers

TL;DR

This paper develops an active learning algorithm that effectively combines strong and weak labelers, reducing label query costs while maintaining high classification accuracy, especially in settings like medical image classification.

Contribution

It introduces a novel active learning framework that leverages both strong and weak labelers, with theoretical guarantees and analysis of label complexity savings.

Findings

The algorithm is statistically consistent.

It reduces label queries compared to using only strong labelers.

Provides conditions for label savings with weak labelers.

Abstract

An active learner is given a hypothesis class, a large set of unlabeled examples and the ability to interactively query labels to an oracle of a subset of these examples; the goal of the learner is to learn a hypothesis in the class that fits the data well by making as few label queries as possible. This work addresses active learning with labels obtained from strong and weak labelers, where in addition to the standard active learning setting, we have an extra weak labeler which may occasionally provide incorrect labels. An example is learning to classify medical images where either expensive labels may be obtained from a physician (oracle or strong labeler), or cheaper but occasionally incorrect labels may be obtained from a medical resident (weak labeler). Our goal is to learn a classifier with low error on data labeled by the oracle, while using the weak labeler to reduce the number of label queries made to this labeler. We provide an active learning algorithm for this setting, establish its statistical consistency, and analyze its label complexity to characterize when it can provide label savings over using the strong labeler alone.