Deep Active Learning for Text Classification with Diverse Interpretations

TL;DR

This paper introduces ALDEN, a novel active learning method for text classification that leverages diverse local interpretations of DNNs to select the most informative samples, reducing annotation costs.

Contribution

ALDEN uniquely uses local interpretability to identify diverse sample regions, improving active learning efficiency in text classification tasks.

Findings

ALDEN outperforms existing active learning methods in experiments.

Using diverse local interpretations enhances sample selection.

The approach effectively reduces labeling effort while maintaining high accuracy.

Abstract

Recently, Deep Neural Networks (DNNs) have made remarkable progress for text classification, which, however, still require a large number of labeled data. To train high-performing models with the minimal annotation cost, active learning is proposed to select and label the most informative samples, yet it is still challenging to measure informativeness of samples used in DNNs. In this paper, inspired by piece-wise linear interpretability of DNNs, we propose a novel Active Learning with DivErse iNterpretations (ALDEN) approach. With local interpretations in DNNs, ALDEN identifies linearly separable regions of samples. Then, it selects samples according to their diversity of local interpretations and queries their labels. To tackle the text classification problem, we choose the word with the most diverse interpretations to represent the whole sentence. Extensive experiments demonstrate that ALDEN consistently outperforms several state-of-the-art deep active learning methods.