新研究解释了优化器为何在等变网络上表现不佳

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-10 04:00

研究人员确定了某些优化器（如 Muon）在训练等变神经网络时优于 Adam 的一个关键原因。问题源于 Adam 在等变线性层的不同块之间处理学习率的方式。具体来说，Adam 会单独重新缩放权重，而不考虑这些块的边界，从而导致步长不一致。提出的解决方案包括单独对每个块的更新进行归一化，并结合调整后的动量系数，这使得 Adam 在各种数据集上能够与 Muon 相媲美。 AI

影响识别出等变网络中一个特定的训练挑战，可能导致更高效的模型开发。

排序理由学术论文，详细介绍了关于神经网络训练的新发现。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 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: 12 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 学术论文，详细介绍了关于神经网络训练的新发现。[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, infra; 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: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Andrei Manolache, Mathias Niepert · 2026-09-10 04:00

Equivariance 破坏了学习率

arXiv:2609.08381v1 Announce Type: cross Abstract: Equivariant networks are commonly trained with Adam, yet recent work reports that matrix-structured optimizers such as Muon can perform better on these architectures without explaining why. We identify one source of this differenc…

本文如何被排名

报道来源 [1]

Equivariance 破坏了学习率

相关实体

相关话题