Enregistré dans:
Détails bibliographiques
Auteurs principaux: Seung, Hyunseok, Lee, Jaewoo, Ko, Hyunsuk
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2506.08353
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913887380570112
author Seung, Hyunseok
Lee, Jaewoo
Ko, Hyunsuk
author_facet Seung, Hyunseok
Lee, Jaewoo
Ko, Hyunsuk
contents We introduce AdaAct, a novel optimization algorithm that adjusts learning rates according to activation variance. Our method enhances the stability of neuron outputs by incorporating neuron-wise adaptivity during the training process, which subsequently leads to better generalization -- a complementary approach to conventional activation regularization methods. Experimental results demonstrate AdaAct's competitive performance across standard image classification benchmarks. We evaluate AdaAct on CIFAR and ImageNet, comparing it with other state-of-the-art methods. Importantly, AdaAct effectively bridges the gap between the convergence speed of Adam and the strong generalization capabilities of SGD, all while maintaining competitive execution times. Code is available at https://github.com/hseung88/adaact.
format Preprint
id arxiv_https___arxiv_org_abs_2506_08353
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle An Adaptive Method Stabilizing Activations for Enhanced Generalization
Seung, Hyunseok
Lee, Jaewoo
Ko, Hyunsuk
Machine Learning
Computer Vision and Pattern Recognition
We introduce AdaAct, a novel optimization algorithm that adjusts learning rates according to activation variance. Our method enhances the stability of neuron outputs by incorporating neuron-wise adaptivity during the training process, which subsequently leads to better generalization -- a complementary approach to conventional activation regularization methods. Experimental results demonstrate AdaAct's competitive performance across standard image classification benchmarks. We evaluate AdaAct on CIFAR and ImageNet, comparing it with other state-of-the-art methods. Importantly, AdaAct effectively bridges the gap between the convergence speed of Adam and the strong generalization capabilities of SGD, all while maintaining competitive execution times. Code is available at https://github.com/hseung88/adaact.
title An Adaptive Method Stabilizing Activations for Enhanced Generalization
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.08353