ZENITH: Automated Gradient Norm Informed Stochastic Optimization

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteur principal: Saha, Dhrubo
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912838436519936
author Saha, Dhrubo
author_facet Saha, Dhrubo
contents Training deep computer vision models requires manual oversight or hyperparameter tuning of the learning rate (LR) schedule. While existing adaptive optimizers schedule the LR automatically, they suffer from computational and memory overhead, incompatibility with regularization, and suboptimal LR choices. In this work, we introduce the ZENITH (Zero-overhead Evolution using Norm-Informed Training History) optimizer, which adapts the LR using the temporal evolution of the gradient norm. Image classification experiments spanning 6 CNN architectures and 6 benchmarks demonstrate that ZENITH achieves higher test accuracy in lower wall-clock time than baselines. It also yielded superior mAP in object detection, keypoint detection, and instance segmentation on MS COCO using the R-CNN family of models. Furthermore, its compatibility with regularization enables even better generalization.
format Preprint
id arxiv_https___arxiv_org_abs_2601_15212
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ZENITH: Automated Gradient Norm Informed Stochastic Optimization
Saha, Dhrubo
Machine Learning
Computer Vision and Pattern Recognition
Training deep computer vision models requires manual oversight or hyperparameter tuning of the learning rate (LR) schedule. While existing adaptive optimizers schedule the LR automatically, they suffer from computational and memory overhead, incompatibility with regularization, and suboptimal LR choices. In this work, we introduce the ZENITH (Zero-overhead Evolution using Norm-Informed Training History) optimizer, which adapts the LR using the temporal evolution of the gradient norm. Image classification experiments spanning 6 CNN architectures and 6 benchmarks demonstrate that ZENITH achieves higher test accuracy in lower wall-clock time than baselines. It also yielded superior mAP in object detection, keypoint detection, and instance segmentation on MS COCO using the R-CNN family of models. Furthermore, its compatibility with regularization enables even better generalization.
title ZENITH: Automated Gradient Norm Informed Stochastic Optimization
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.15212