Branch Scaling Manifests as Implicit Architectural Regularization for Improving Generalization in Overparameterized ResNets
Abstract
Scaling factors in residual branches have emerged as a prevalent method for boosting neural network performance, especially in normalization-free architectures. While prior work has primarily examined scaling effects from an optimization perspective, this paper investigates their role in residual architectures through the lens of generalization theory. Specifically, we establish that wide residual networks (ResNets) with constant scaling factors become asymptotically unlearnable as depth increases. In contrast, when the scaling factor exhibits rapid depth-wise decay combined with early stopping, over-parameterized ResNets achieve minimax-optimal generalization rates. To establish this, we demonstrate that the generalization capability of wide ResNets can be approximated by kernel regression associated with the Neural Tangent Kernel (NTK). Our theoretical findings are validated through experiments on synthetic data and real-world classification tasks, including MNIST and CIFAR-100.
Cite
@article{arxiv.2403.04545,
title = {Branch Scaling Manifests as Implicit Architectural Regularization for Improving Generalization in Overparameterized ResNets},
author = {Zixiong Yu and Guhan Chen and Jianfa Lai and Bohan Li and Songtao Tian},
journal= {arXiv preprint arXiv:2403.04545},
year = {2026}
}
Comments
Accepted by ICML. This version incorporates content from the preprint arXiv:2305.18506. The contributors of the relevant content have consented to its inclusion and have been listed as authors