What-If Explanations Over Time: Counterfactuals for Time Series Classification

TL;DR

This paper surveys recent algorithms for counterfactual explanations in time series classification, highlighting challenges, evaluating methods, and providing an open-source library for standardization and practical use.

Contribution

It offers a comprehensive review of state-of-the-art counterfactual methods for time series, discusses unique temporal challenges, and introduces an open-source framework for evaluation.

Findings

Analyzed strengths and limitations of existing approaches.

Compared effectiveness along key dimensions like validity and plausibility.

Provided an open-source library to standardize evaluation.

Abstract

Counterfactual explanations emerge as a powerful approach in explainable AI, providing what-if scenarios that reveal how minimal changes to an input time series can alter the model's prediction. This work presents a survey of recent algorithms for counterfactual explanations for time series classification. We review state-of-the-art methods, spanning instance-based nearest-neighbor techniques, pattern-driven algorithms, gradient-based optimization, and generative models. For each, we discuss the underlying methodology, the models and classifiers they target, and the datasets on which they are evaluated. We highlight unique challenges in generating counterfactuals for temporal data, such as maintaining temporal coherence, plausibility, and actionable interpretability, which distinguish the temporal from tabular or image domains. We analyze the strengths and limitations of existing approaches and compare their effectiveness along key dimensions (validity, proximity, sparsity, plausibility, etc.). In addition, we implemented an open-source implementation library, Counterfactual Explanations for Time Series (CFTS), as a reference framework that includes many algorithms and evaluation metrics. We discuss this library's contributions in standardizing evaluation and enabling practical adoption of explainable time series techniques. Finally, based on the literature and identified gaps, we propose future research directions, including improved user-centered design, integration of domain knowledge, and counterfactuals for time series forecasting.