TimeSliver : Symbolic-Linear Decomposition for Explainable Time Series Classification

TL;DR

TimeSliver is a novel explainability framework for time series classification that combines raw data and symbolic abstraction to produce interpretable importance scores for each temporal segment, outperforming existing methods.

Contribution

It introduces a new symbolic-linear decomposition approach that enhances interpretability and maintains temporal structure, improving attribution accuracy and predictive performance.

Findings

Outperforms other attribution methods by 11% on 7 datasets.

Achieves within 2% of state-of-the-art accuracy on 26 datasets.

Provides meaningful importance scores for temporal segments.

Abstract

Identifying the extent to which every temporal segment influences a model's predictions is essential for explaining model decisions and increasing transparency. While post-hoc explainable methods based on gradients and feature-based attributions have been popular, they suffer from reference state sensitivity and struggle to generalize across time-series datasets, as they treat time points independently and ignore sequential dependencies. Another perspective on explainable time-series classification is through interpretable components of the model, for instance, leveraging self-attention mechanisms to estimate temporal attribution; however, recent findings indicate that these attention weights often fail to provide faithful measures of temporal importance. In this work, we advance this perspective and present a novel explainability-driven deep learning framework, TimeSliver, which jointly utilizes raw time-series data and its symbolic abstraction to construct a representation that maintains the original temporal structure. Each element in this representation linearly encodes the contribution of each temporal segment to the final prediction, allowing us to assign a meaningful importance score to every time point. For time-series classification, TimeSliver outperforms other temporal attribution methods by 11% on 7 distinct synthetic and real-world multivariate time-series datasets. TimeSliver also achieves predictive performance within 2% of state-of-the-art baselines across 26 UEA benchmark datasets, positioning it as a strong and explainable framework for general time-series classification.