Abstract

Various techniques focus on understanding the way an image or a category concept is represented within convolutional neural networks. It is common to assume that one latent neuron can be a detector for a category or its dominant features. The analysis of the collective activity of neurons in hidden layers shows that representation of most of the categories is complex and distributed across as many as 93% of units participating in the encoding. To account for the complexity, this study suggests an approach that represents category by constructing a prototype formed as a covariance matrix of layer neurons’ activation. The approach makes it possible to take into account the coordinated response of the population of artificial neurons, as well as to preserve the complexity of the distributed activation pattern at different stages of processing, from low-level image statistics to the level of abstract and semantic features.

© 2022 Optica Publishing Group

Processing of chromatic information in a deep convolutional neural network

Alban Flachot and Karl R. Gegenfurtner
J. Opt. Soc. Am. A 35(4) B334-B346 (2018)

Convolutional Neural Network-Based Optical Performance Monitoring for Optical Transport Networks

Takahito Tanimura, Takeshi Hoshida, Tomoyuki Kato, Shigeki Watanabe, and Hiroyuki Morikawa
J. Opt. Commun. Netw. 11(1) A52-A59 (2019)

Backpropagation through nonlinear units for the all-optical training of neural networks

Xianxin Guo, Thomas D. Barrett, Zhiming M. Wang, and A. I. Lvovsky
Photon. Res. 9(3) B71-B80 (2021)

Deep attentive convolutional neural network for automatic grading of imbalanced diabetic retinopathy in retinal fundus images

Feng Li, Shiqing Tang, Yuyang Chen, and Haidong Zou
Biomed. Opt. Express 13(11) 5813-5835 (2022)

High-accuracy optical convolution unit architecture for convolutional neural networks by cascaded acousto-optical modulator arrays

Shaofu Xu, Jing Wang, Rui Wang, Jiangping Chen, and Weiwen Zou
Opt. Express 27(14) 19778-19787 (2019)