Deep Active Learning by Model Interpretability

TL;DR

This paper introduces DAMI, a novel deep active learning method leveraging model interpretability through piece-wise linear regions in DNNs, to efficiently select informative samples and reduce annotation costs.

Contribution

DAMI uniquely uses piece-wise linear interpretability in DNNs to guide sample selection in active learning without hyper-parameter tuning.

Findings

DAMI outperforms state-of-the-art active learning methods on tabular datasets.

The approach effectively identifies representative samples using interpretability-based clustering.

No hyper-parameter tuning required for DAMI, simplifying its application.

Abstract

Recent successes of Deep Neural Networks (DNNs) in a variety of research tasks, however, heavily rely on the large amounts of labeled samples. This may require considerable annotation cost in real-world applications. Fortunately, active learning is a promising methodology to train high-performing model with minimal annotation cost. In the deep learning context, the critical question of active learning is how to precisely identify the informativeness of samples for DNN. In this paper, inspired by piece-wise linear interpretability in DNN, we introduce the linearly separable regions of samples to the problem of active learning, and propose a novel Deep Active learning approach by Model Interpretability (DAMI). To keep the maximal representativeness of the entire unlabeled data, DAMI tries to select and label samples on different linearly separable regions introduced by the piece-wise linear interpretability in DNN. We focus on modeling Multi-Layer Perception (MLP) for modeling tabular data. Specifically, we use the local piece-wise interpretation in MLP as the representation of each sample, and directly run K-Center clustering to select and label samples. To be noted, this whole process of DAMI does not require any hyper-parameters to tune manually. To verify the effectiveness of our approach, extensive experiments have been conducted on several tabular datasets. The experimental results demonstrate that DAMI constantly outperforms several state-of-the-art compared approaches.