English(EN) SHANG++: Robust Stochastic Acceleration under Multiplicative Noise

新的SHANG++方法增强了AI模型在噪声下的训练鲁棒性

作者 PulseAugur 编辑部 · [1 个来源] · 2026-08-20 04:00

研究人员开发了两种新的加速随机梯度下降方法，SHANG和SHANG++，旨在提高在乘性噪声下的稳定性和收敛性。SHANG是一种半隐式离散化方法，可增强稳定性，而SHANG++则包含阻尼校正，以实现更快的收敛和更高的噪声鲁棒性。这两种方法都被证明对凸目标和强凸目标有效，SHANG++在各种应用中都表现出一致的性能，包括深度学习任务，如训练ResNet-34，在该任务中，其准确性接近无噪声设置。 AI

影响提高了深度学习模型的训练效率和鲁棒性，可能使在噪声数据下进行更稳定的训练成为可能。

排序理由详细介绍机器学习新优化方法的学术论文。[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; 详细介绍机器学习新优化方法的学术论文。[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: 49 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) · Yaxin Yu, Long Chen, Minfu Feng · 2026-08-20 04:00

SHANG++：多重噪声下的鲁棒随机加速

arXiv:2603.09355v2 Announce Type: replace-cross Abstract: Under the multiplicative noise scaling (MNS) condition, original Nesterov acceleration is provably sensitive to noise and may diverge when gradient noise overwhelms the signal. In this paper, we develop two accelerated sto…

本文如何被排名

报道来源 [1]

SHANG++：多重噪声下的鲁棒随机加速

相关实体

相关话题