A shallow residual neural network to predict the visual cortex response

TL;DR

This paper demonstrates that a shallow residual neural network can effectively predict human visual cortex responses, with improved accuracy over previous models, by enabling better training of early network stages.

Contribution

It introduces a shallow residual neural network architecture for predicting visual cortex activity, improving training and prediction accuracy compared to prior models.

Findings

Prediction accuracy increased from 10.4% to 15.53%.

Additional training epochs further enhance performance.

Residual connections facilitate training of earlier network layers.

Abstract

Understanding how the visual cortex of the human brain really works is still an open problem for science today. A better understanding of natural intelligence could also benefit object-recognition algorithms based on convolutional neural networks. In this paper we demonstrate the asset of using a shallow residual neural network for this task. The benefit of this approach is that earlier stages of the network can be accurately trained, which allows us to add more layers at the earlier stage. With this additional layer the prediction of the visual brain activity improves from $10.4\%$ (block 1) to $15.53\%$ (last fully connected layer). By training the network for more than 10 epochs this improvement can become even larger.