Anatomically Constrained ResNets Exhibit Opponent Receptive Fields; So What?

Ethan Harris, Daniela Mihai, Jonathon Hare

Keywords: bottlenecks, opponency, adversarial robustness, shape bias

Abstract: Primate visual systems are well known to exhibit varying degrees of bottlenecks in the early visual pathway. Recent works have shown that the presence of a bottleneck between 'retinal' and 'ventral' parts of artificial models of visual systems, simulating the optic nerve, can cause the emergence of cellular properties that have been observed in primates: namely centre-surround organisation and opponency. To date, however, state-of-the-art convolutional network architectures for classification problems have not incorporated such an early bottleneck. In this paper, we ask what happens if such a bottleneck is added to a ResNet-50 model trained to classify the ImageNet data set. Our experiments show that some of the emergent properties observed in simpler models still appear in these considerably deeper and more complex models, however, there are some notable differences particularly with regard to spectral opponency. The introduction of the bottleneck is experimentally shown to introduce a small but consistent shape bias into the network. Tight bottlenecks are also shown to only have a very slight affect on the top-1 accuracy of the models when trained and tested on ImageNet.

TL;DR: We ask what happens when a bottleneck is added to a ResNet-50 trained on ImageNet