Beyond Sufficiency: Time Series Explanation with Counterfactual Necessity
2026-07-23 • Machine Learning
Machine LearningArtificial Intelligence
AI summaryⓘ
The authors point out that current methods explaining time-series classifiers often highlight parts of the data that support a prediction but aren’t truly necessary for it. They introduce TimePNS, a new method that checks if parts of the time series are really needed by testing what happens when those parts are changed or removed. Their approach works in two steps: first, it finds important segments that seem to explain the prediction, then it refines this by seeing which segments actually affect the outcome when altered. Tests show that TimePNS does better at finding the genuinely important parts in time-series data compared to older methods.
time-series classificationexplainable AIcounterfactual interventioncausal inferencesufficiencynecessityblack-box modeltemporal factorscausal generative processexplanation mask
Authors
Hongnan Ma, Yiwei Shi, Mengyue Yang, Weiru Liu
Abstract
Faithful explanations of time-series classifiers should identify subsequences that are not only sufficient to preserve a black-box model's prediction, but also necessary for maintaining it. However, existing sufficiency-oriented methods can assign high importance to spurious subsequences that support the prediction without being essential to the model's decision. We introduce \textbf{TimePNS}, a necessity-aware framework for time-series explanation. Inspired by Pearl's counterfactual notion of necessity, TimePNS assesses whether a temporal factor is necessary by intervening on it and measuring whether the original prediction is disrupted. The framework adopts a two-stage design. Stage I learns an identifiable causal generative process together with a sufficiency-oriented explanation mask. Stage II performs counterfactual interventions on temporal factors to derive necessity signals, which supervise a temporal gate that refines the initial explanation by suppressing non-essential components and emphasizing counterfactually necessary ones. Experiments on synthetic and real-world time-series benchmarks show that TimePNS more accurately identifies decision-critical subsequences and consistently improves sufficiency-necessity trade-offs over strong baselines.