FLEXtime: Filterbank learning to explain time series
This addresses the challenge of interpreting complex time series data, such as EEG and audio, for researchers and practitioners in fields like signal processing and machine learning, though it is incremental by building on established signal decomposition techniques.
The authors tackled the problem of explaining predictions from time series by proposing FLEXtime, a method that uses a bank of bandpass filters to create saliency maps over frequency bands, which outperformed state-of-the-art explainability methods on average across multiple datasets.
State-of-the-art methods for explaining predictions from time series involve learning an instance-wise saliency mask for each time step; however, many types of time series are difficult to interpret in the time domain, due to the inherently complex nature of the data. Instead, we propose to view time series explainability as saliency maps over interpretable parts, leaning on established signal processing methodology on signal decomposition. Specifically, we propose a new method called FLEXtime that uses a bank of bandpass filters to split the time series into frequency bands. Then, we learn the combination of these bands that optimally explains the model's prediction. Our extensive evaluation shows that, on average, FLEXtime outperforms state-of-the-art explainability methods across a range of datasets. FLEXtime fills an important gap in the current time series explainability methodology and is a valuable tool for a wide range of time series such as EEG and audio. Code is available at https://github.com/theabrusch/FLEXtime.