Inherently Interpretable Time Series Classification via Multiple Instance Learning

TL;DR

This paper introduces MILLET, a framework that makes deep learning time series classifiers inherently interpretable using Multiple Instance Learning, achieving high-quality explanations without sacrificing accuracy across diverse datasets.

Contribution

The paper develops the first general MIL-based approach for interpretable time series classification, applicable to various models and datasets.

Findings

MILLET produces sparse, high-quality explanations.

It maintains or improves predictive performance.

It is validated on 85 UCR datasets and a synthetic dataset.

Abstract

Conventional Time Series Classification (TSC) methods are often black boxes that obscure inherent interpretation of their decision-making processes. In this work, we leverage Multiple Instance Learning (MIL) to overcome this issue, and propose a new framework called MILLET: Multiple Instance Learning for Locally Explainable Time series classification. We apply MILLET to existing deep learning TSC models and show how they become inherently interpretable without compromising (and in some cases, even improving) predictive performance. We evaluate MILLET on 85 UCR TSC datasets and also present a novel synthetic dataset that is specially designed to facilitate interpretability evaluation. On these datasets, we show MILLET produces sparse explanations quickly that are of higher quality than other well-known interpretability methods. To the best of our knowledge, our work with MILLET, which is available on GitHub (https://github.com/JAEarly/MILTimeSeriesClassification), is the first to develop general MIL methods for TSC and apply them to an extensive variety of domains