What are there the best ways to leverage the unique "geometrical" constraints of spectro-temporal signal representations (architecture, filter shapes, data augmentation, etc.)? For example, a learned feature in a spectrogram should intuitively be horizontally (temporally) invariant since the window of time when it was captured is arbitrary. In contrast, the specific vertical position of a feature is extremely important since the precise frequency at which it occurred is significant. My research group is trying do design a convolutional neural network (using keras/tensorflow) that takes spectrograms representing neural data and predicts a particular disease biomarker. All of our trained models so far severely overfit and we have somewhat limited data.

Just for further background:

We decided to use CNNs because the visual representations of spectral data are known to be useful for our task. Even though our data set is not huge, we were hoping that CNNs could be more interpretable in terms of exactly what spectral features are being learned than other ML methods like SVMs or LSTMs. The models have been regressors so far, but we are open to switching to classifiers if they perform better in the end.

  • $\begingroup$ Do you mean that some neural signal shows up intermittently and so you want to use the CNN whether the important part of the signal shows up anywhere in the measurement? $\endgroup$ – Engineer Jul 7 at 22:23
  • $\begingroup$ @Engineer Yes, that sounds right if I understand you correctly. I'm just not sure exactly what you mean by the "important part" of the signal. If we can identify any spectral characteristic(s) of the neural signal that modulate the biomarker by visualizing the learned features of the CNN, we would be happy! $\endgroup$ – bez Jul 8 at 23:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.