cs.AI, cs.LG, math.ST, stat.TH

Path Regularization: A Near-Complete and Optimal Nonasymptotic Generalization Theory for Multilayer Neural Networks and Double Descent Phenomenon

arXiv:2503.02129v2 Announce Type: replace-cross
Abstract: Path regularization has shown to be a very effective regularization to train neural networks, leading to a better generalization property than common regularizations i.e. weight decay, etc. We …