Selectivity considered harmful: evaluating the causal impact of class selectivity in DNNs

Posted 4 months ago
Website Profile:
Computers & Internet

It remains an open question whether it is necessary and/or sufficient for deep neural networks (DNNs) to learn class selectivity in individual units.

We investigated the causal impact of class selectivity on network function by directly regularizing for or against class selectivity. Using this regularizer to reduce class selectivity across units in convolutional neural networks increased test accuracy by over 2% for ResNet18 trained on Tiny ImageNet.

Read the whole paper here. Download this paper here.