FreqRISE: Explaining time series using frequency masking

TL;DR

FreqRISE introduces a novel frequency domain masking approach to generate explanations for time series data, emphasizing the importance of frequency-based saliency over raw input localization for improved interpretability.

Contribution

It proposes FreqRISE, a new method that explains time series by masking in the frequency and time-frequency domains, outperforming existing saliency methods.

Findings

Outperforms strong baselines in multiple tasks

Highlights the importance of frequency domain explanations

Provides accessible source code for reproducibility

Abstract

Time-series data are fundamentally important for many critical domains such as healthcare, finance, and climate, where explainable models are necessary for safe automated decision making. To develop explainable artificial intelligence in these domains therefore implies explaining salient information in the time series. Current methods for obtaining saliency maps assume localized information in the raw input space. In this paper, we argue that the salient information of a number of time series is more likely to be localized in the frequency domain. We propose FreqRISE, which uses masking-based methods to produce explanations in the frequency and time-frequency domain, and outperforms strong baselines across a number of tasks. The source code is available here: \url{https://github.com/theabrusch/FreqRISE}.