Gradient Regularization Improves Accuracy of Discriminative Models

Varga, Dániel and Csiszárik, Adrián and Zombori, Zsolt (2018) Gradient Regularization Improves Accuracy of Discriminative Models. Schedae Informaticae, 27. pp. 31-45. ISSN 0860-0295

Preview

Text 10408.pdf Download (161kB) | Preview

Abstract

Regularizing the gradient norm of the output of a neural network is a powerful technique, rediscovered several times. This paper presents evidence that gradient regularization can consistently improve classification accuracy on vision tasks, using modern deep neural networks, especially when the amount of training data is small. We introduce our regularizers as members of a broader class of Jacobian-based regularizers. We demonstrate empirically on real and synthetic data that the learning process leads to gradients controlled beyond the training points, and results in solutions that generalize well.

Actions (login required)

Edit Item