cs.LGNov 7, 2026

Predictive Surprise as Self-Grounding Concept Bottleneck for Interpretable Time Series

Authors: Sachith Abeywickrama, Emadeldeen Eldele, Min Wu, Xiaoli Li, Chau Yuen

Organizations: Nanyang Technological University, Singapore · Institute of Advanced Intelligence and Computing (IAIC), A*STAR, Singapore · Department of Computer Science, Khalifa University, UAE · Singapore University of Technology and Design, Singapore

Abstract

Time-series classifiers deployed in risk-sensitive domains must be (1) accurate, (2) interpretable, and (3) able to abstain when evidence is weak. Existing methods do not deliver all 3 at once, and the obstacle is structural, rather than logistical. Meaningful temporal patterns cannot be defined without knowing segment boundaries, yet meaningful boundaries cannot be placed without knowing which patterns to expect. Existing approaches break this circularity by using fixed windows or expert-annotated vocabularies, thereby sacrificing at least one of the 3 properties. We introduce ConceptTime, which closes the loop through a single self-supervised signal. A frozen probabilistic forecaster predicts a distribution over the next observation at every timestep of the sequence. We call the mismatch between its prediction and the realized signal the predictive surprise. This signal both locates segment boundaries and characterizes each resulting segment through the forecaster. Segments are thus grounded by how they behaved relative to expectation, not by what they look like. These summaries cluster into a vocabulary of dynamical regimes that we call concepts. A lightweight head classifies from the concept sequence, and the distance to the nearest concept yields reliability by design. Across 17 public benchmarks, ConceptTime is competitive with state-of-the-art black-box accuracy, beats every interpretable baseline, and retains near-98% accuracy on Epilepsy with 1% of labels. It further remains competitive with softmax, prototype-distance, and SHAP/TimeX/TimeX++ on OOD detection and deletion faithfulness, without a single annotation, language model call, or domain expert. Code is available at: https://github.com/Sachithx/ConceptTime.git.

Figures & tables

Appendix figures & tables19 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. ConceptTS: LLM-Guided Concept Bottlenecks for Interpretable Multivariate Time-Series Forecasting

    Aug 21, 2026Yichen Jiang, Yueqiao Chen, Dongyu LiuMultivariate Time Series ForecastingExplainable Time Series Forecasting

  2. Fast, Interpretable, and Deterministic Time Series Classification With a Bag-of-Receptive-Fields

    Nov 29, 2023Francesco Spinnato, Riccardo Guidotti, Anna Monreale +1Time Series ClassificationInterpretable ML

  3. Beyond Sufficiency: Time Series Explanation with Counterfactual Necessity

    Jul 23, 2026Hongnan Ma, Yiwei Shi, Mengyue Yang +1Counterfactual ExplanationsTime Series Classification