Regularizing activations in neural networks via distribution matching with the Wasserstein metric
This work addresses the need for better regularization techniques in deep learning, though it appears incremental as it builds on existing distribution matching concepts.
The authors tackled the problem of improving neural network training and generalization by proposing a new regularization method that encourages activations to follow a standard normal distribution, achieving competitive performance on image classification and language modeling tasks.
Regularization and normalization have become indispensable components in training deep neural networks, resulting in faster training and improved generalization performance. We propose the projected error function regularization loss (PER) that encourages activations to follow the standard normal distribution. PER randomly projects activations onto one-dimensional space and computes the regularization loss in the projected space. PER is similar to the Pseudo-Huber loss in the projected space, thus taking advantage of both $L^1$ and $L^2$ regularization losses. Besides, PER can capture the interaction between hidden units by projection vector drawn from a unit sphere. By doing so, PER minimizes the upper bound of the Wasserstein distance of order one between an empirical distribution of activations and the standard normal distribution. To the best of the authors' knowledge, this is the first work to regularize activations via distribution matching in the probability distribution space. We evaluate the proposed method on the image classification task and the word-level language modeling task.