PulseAugur
实时 13:55:49
English(EN) FOGO: Forgetting-aware Orthogonalization Optimizer

新的FOGO优化器解决AI模型遗忘问题

研究人员推出了一种新颖的优化器FOGO,旨在解决AI模型训练过程中的遗忘问题。FOGO通过检测和解决梯度干扰,同时处理每个训练步骤中的短期遗忘和持续学习中常见的长期遗忘。该优化器利用谱正交化和紧凑的代码本记忆来保留过去的更新方向,在包括微调LLaVA-7B和预训练GPT-2在内的各种任务中,均表现出比Adam和Muon等现有优化器更好的收敛性和知识保留能力。 AI

影响 FOGO减少遗忘的能力可能导致更高效、更有效的AI模型训练,尤其是在持续学习场景中。

排序理由 该集群包含一篇详细介绍AI模型新优化算法的研究论文。

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

报道来源 [2]

  1. arXiv cs.AI TIER_1 English(EN) · Toan Nguyen, Yang Liu, Trung Le, Celso de Melo, Flora D. Salim ·

    FOGO: Forgetting-aware Orthogonalization Optimizer

    arXiv:2606.10406v1 Announce Type: cross Abstract: We argue that forgetting is not confined to continual learning but is a general optimization phenomenon: during standard training, dominant mini-batch gradients suppress rare but useful update directions, causing short-term forget…

  2. Hugging Face Daily Papers TIER_1 English(EN) ·

    FOGO: Forgetting-aware Orthogonalization Optimizer

    We argue that forgetting is not confined to continual learning but is a general optimization phenomenon: during standard training, dominant mini-batch gradients suppress rare but useful update directions, causing short-term forgetting at every step. When such knowledge is never r…