Estimating the Generalization in Deep Neural Networks via Sparsity
This addresses the challenge of reliably measuring generalization ability in DNNs, which is crucial for model evaluation and deployment, though it appears incremental as it builds on existing sparsity concepts.
The paper tackles the problem of estimating the generalization gap in deep neural networks by proposing a novel method based on network sparsity, using two key quantities and a linear model, and demonstrates its efficiency on popular datasets.
Generalization is the key capability for deep neural networks (DNNs). However, it is challenging to give a reliable measure of the generalization ability of a DNN via only its nature. In this paper, we propose a novel method for estimating the generalization gap based on network sparsity. In our method, two key quantities are proposed first. They have close relationship with the generalization ability and can be calculated directly from the training results alone. Then a simple linear model involving two key quantities are constructed to give accurate estimation of the generalization gap. By training DNNs with a wide range of generalization gap on popular datasets, we show that our key quantities and linear model could be efficient tools for estimating the generalization gap of DNNs.