Information-Theoretic Understanding of Population Risk Improvement with Model Compression

arXiv1901.09421MaRDI QIDQ6313131

Author name not available (Why is that?)

Publication date: 27 January 2019

Abstract: We show that model compression can improve the population risk of a pre-trained model, by studying the tradeoff between the decrease in the generalization error and the increase in the empirical risk with model compression. We first prove that model compression reduces an information-theoretic bound on the generalization error; this allows for an interpretation of model compression as a regularization technique to avoid overfitting. We then characterize the increase in empirical risk with model compression using rate distortion theory. These results imply that the population risk could be improved by model compression if the decrease in generalization error exceeds the increase in empirical risk. We show through a linear regression example that such a decrease in population risk due to model compression is indeed possible. Our theoretical results further suggest that the Hessian-weighted $K$-means clustering compression approach can be improved by regularizing the distance between the clustering centers. We provide experiments with neural networks to support our theoretical assertions.

Has companion code repository: https://github.com/wgao9/weight_quant

This page was built for publication: Information-Theoretic Understanding of Population Risk Improvement with Model Compression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6313131)