Rethinking Saliency Map: An Context-aware Perturbation Method to Explain EEG-based Deep Learning Model
This addresses the lack of appropriate explanation methods for EEG-based deep learning models, which is an incremental improvement for researchers in brain-computer interfaces and neuroscience.
The authors tackled the problem of explaining EEG-based deep learning models by proposing a context-aware perturbation method to generate saliency maps from raw EEG signals, which also suppresses artifacts; experiments on the DEAP dataset showed advantages over other methods.
Deep learning is widely used to decode the electroencephalogram (EEG) signal. However, there are few attempts to specifically investigate how to explain the EEG-based deep learning models. We conduct a review to summarize the existing works explaining the EEG-based deep learning model. Unfortunately, we find that there is no appropriate method to explain them. Based on the characteristic of EEG data, we suggest a context-aware perturbation method to generate a saliency map from the perspective of the raw EEG signal. Moreover, we also justify that the context information can be used to suppress the artifacts in the EEG-based deep learning model. In practice, some users might want a simple version of the explanation, which only indicates a few features as salient points. To this end, we propose an optional area limitation strategy to restrict the highlighted region. To validate our idea and make a comparison with the other methods, we select three representative EEG-based models to implement experiments on the emotional EEG dataset DEAP. The results of the experiments support the advantages of our method.