English(EN) Fix the Loss, Not the Radius: Rethinking the Adversarial Perturbation of Sharpness-Aware Minimization

新的 LE-SAM 方法在模型泛化方面优于传统 SAM

作者 PulseAugur 编辑部 · [1 个来源] · 2026-05-11 08:34

研究人员推出了一种名为损失均衡 SAM (LE-SAM) 的新方法，以增强机器学习模型的泛化能力。该方法通过关注固定的损失空间预算而非固定的扰动半径，解决了锐度感知最小化 (SAM) 中的不匹配问题。LE-SAM 有效地优先考虑了以曲率为导向的优化项而非梯度范数信号。实验表明，LE-SAM 在各种基准测试中持续优于 SAM 及其变体，取得了最先进的成果。 AI

影响引入了一种新的优化技术，可提高模型泛化能力，有望带来更强大的 AI 系统。

排序理由介绍机器学习优化新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 介绍机器学习优化新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, other; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 150 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Zhiqiang Gao · 2026-05-11 08:34

修复损失而非半径：重新思考锐度感知最小化的对抗性扰动

Sharpness-Aware Minimization (SAM) improves generalization by minimizing the worst-case loss within a fixed parameter-space radius neighborhood. SAM and its variants mainly rely on a first-order linearized surrogate, while flat minima are inherently a second-order (curvature) not…

本文如何被排名

报道来源 [1]

修复损失而非半径：重新思考锐度感知最小化的对抗性扰动

相关实体

相关话题