We are introducing a saliency map method that produces meaningful attributions for time-series ML models. Instead of restricting explanations to the time domain, our approach generates saliency maps in relevant domains such as frequency or ICA components - improving interpretability when applying deep models to EEG/MEG/LFP.

Cross-Domain Saliency Maps is an open-source toolkit that:
  • Provides frequency and ICA domain attributions out-of-the box
  • Extends to any invertible transform with a differentiable inverse
  • Works plug-and-play with PyTorch and TensorFlow, no retraining required
  • Demonstrates utility on EEG seizure detection and other datasets

Get started:

Get the code & run it on your data (GitHub repo)
https://github.com/esl-epfl/cross-domain-saliency-maps 

Read the full story (arXiv preprint)
https://arxiv.org/pdf/2505.13100

We would be happy to hear your thoughts and experiences applying this to neural data.