Inherently Interpretable Time Series Classification via Multiple Instance Learning

Joseph Early, Gavin K. C. Cheung, Kurt Cutajar, Hanting Xie, Jas Kandola +1 more
2/12/2026

Official code: JAEarly/MILTimeSeriesClassification (57 stars, Apache-2.0, last commit 2024-03-16).

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

View on arXivView PDF

Code Implementations(1)

JAEarly/MILTimeSeriesClassificationOfficial100%
Apache-2.0

Inherently Interpretable Time Series Classification via Multiple Instance Learning (MILLET)

5713Python, Jupyter NotebookNov 16, 20232 years agoApache-2.0

Similar Papers

L2GTX: From Local to Global Time Series Explanations

87% match2026

XTSC-Bench: Quantitative Benchmarking for Explainers on Time Series Classification

84% match1 repo42023

Interactive Counterfactual Generation for Univariate Time Series

83% match1 repo42024

Improving the Evaluation and Actionability of Explanation Methods for Multivariate Time Series Classification

82% match2 repos92024

Robust Explainer Recommendation for Time Series Classification

82% match1 repo72023

EXCODER: EXplainable Classification Of DiscretE time series Representations

82% match2026

Cite this paper

@article{early2026inherently,
  title  = {Inherently Interpretable Time Series Classification via Multiple Instance Learning},
  author = {Joseph Early and Gavin K. C. Cheung and Kurt Cutajar and Hanting Xie and Jas Kandola and Niall Twomey},
  year   = {2026},
  eprint = {2311.10049},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/2311.10049},
  journal = {ICLR 2023 2023}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2311.10049.svg)](https://codeofpaper.com/paper/2311.10049)

Discussion