This article is cited in 1 paper

Part 1. General problems of the intellectual systems theory

Neural network regularization via SVD for small dataset tasks

N. V. Vaulin

Moscow State Pedagogical University

Abstract: Consider convolutional neural network optimization problem using small dataset. The proposed method fine-tune pretrained neural network with specigic constrains for convolutional kernels. Pretrained weights are decomposed using SVD. During fine-tuning the only singular values are trained. In the paper are investigated the dinamic of singular values during training process and its influence on quality of resulting model. The method are comapared with other approaches in rengenological images classsification problem.

Keywords: deep learning, convolutional neural networks, singular value decomposition.