TimeREISE: Time-series Randomized Evolving Input Sample Explanation

TL;DR

TimeREISE is a model-agnostic explanation method for time series classifiers that offers superior interpretability, efficiency, and applicability without relying on prior data, addressing a key gap in explainable AI for time series.

Contribution

It introduces TimeREISE, a novel attribution method specifically designed for time series classification, improving interpretability and efficiency over existing approaches.

Findings

Outperforms existing methods on standard metrics

Applicable to any time series classification network

Runtime scales non-linearly with input size

Abstract

Deep neural networks are one of the most successful classifiers across different domains. However, due to their limitations concerning interpretability their use is limited in safety critical context. The research field of explainable artificial intelligence addresses this problem. However, most of the interpretability methods are aligned to the image modality by design. The paper introduces TimeREISE a model agnostic attribution method specifically aligned to success in the context of time series classification. The method shows superior performance compared to existing approaches concerning different well-established measurements. TimeREISE is applicable to any time series classification network, its runtime does not scale in a linear manner concerning the input shape and it does not rely on prior data knowledge.