TimeSAE: Sparse Decoding for Faithful Explanations of Black-Box Time Series Models
This addresses the need for interpretable explanations in high-stakes time series applications, offering an incremental improvement over existing methods by enhancing generalization outside training data.
The authors tackled the problem of explaining black-box time series models, which often fail under distributional shifts, by introducing TimeSAE, a framework using Sparse Autoencoders and causality, and showed it provides more faithful and robust explanations compared to baselines in evaluations on synthetic and real-world datasets.
As black box models and pretrained models gain traction in time series applications, understanding and explaining their predictions becomes increasingly vital, especially in high-stakes domains where interpretability and trust are essential. However, most of the existing methods involve only in-distribution explanation, and do not generalize outside the training support, which requires the learning capability of generalization. In this work, we aim to provide a framework to explain black-box models for time series data through the dual lenses of Sparse Autoencoders (SAEs) and causality. We show that many current explanation methods are sensitive to distributional shifts, limiting their effectiveness in real-world scenarios. Building on the concept of Sparse Autoencoder, we introduce TimeSAE, a framework for black-box model explanation. We conduct extensive evaluations of TimeSAE on both synthetic and real-world time series datasets, comparing it to leading baselines. The results, supported by both quantitative metrics and qualitative insights, show that TimeSAE provides more faithful and robust explanations. Our code is available in an easy-to-use library TimeSAE-Lib: https://anonymous.4open.science/w/TimeSAE-571D/.