Stochastic Weight Matrix-based Regularization Methods for Deep Neural Networks
This addresses the problem of regularization for deep learning practitioners, but it is incremental as it builds on existing weight modification techniques.
The paper tackles overfitting in deep neural networks by introducing two regularization methods that directly modify weight matrices, reporting performance gains and increased entropy on benchmark datasets like MNIST and CIFAR-10.
The aim of this paper is to introduce two widely applicable regularization methods based on the direct modification of weight matrices. The first method, Weight Reinitialization, utilizes a simplified Bayesian assumption with partially resetting a sparse subset of the parameters. The second one, Weight Shuffling, introduces an entropy- and weight distribution-invariant non-white noise to the parameters. The latter can also be interpreted as an ensemble approach. The proposed methods are evaluated on benchmark datasets, such as MNIST, CIFAR-10 or the JSB Chorales database, and also on time series modeling tasks. We report gains both regarding performance and entropy of the analyzed networks. We also made our code available as a GitHub repository (https://github.com/rpatrik96/lod-wmm-2019).