On the Generalization of Models Trained with SGD: Information-Theoretic Bounds and Implications
It addresses generalization analysis for machine learning practitioners, but is incremental as it builds on prior work.
The paper presents new information-theoretic upper bounds for the generalization error of models trained with SGD, applying them to analyze linear and two-layer ReLU networks, and introduces a simple regularization scheme that performs comparably to state-of-the-art methods.
This paper follows up on a recent work of Neu et al. (2021) and presents some new information-theoretic upper bounds for the generalization error of machine learning models, such as neural networks, trained with SGD. We apply these bounds to analyzing the generalization behaviour of linear and two-layer ReLU networks. Experimental study of these bounds provide some insights on the SGD training of neural networks. They also point to a new and simple regularization scheme which we show performs comparably to the current state of the art.