CoGraM通过回滚机制提升模型融合稳定性与精度。
CoGraM: Context-sensitive granular optimization method with rollback for robust model fusion
- 分层分粒度迭代优化，根据损失差异动态调整
- 相比Fisher等方法，显著提升融合后模型准确率
- 适合需要稳定模型融合的联邦学习场景
无需重新训练的神经网络合并是联邦与分布式学习的核心。现有方法如权重平均或Fisher合并常导致精度下降且跨种子表现不稳定。CoGraM（上下文感知粒度合并）是一种多阶段、上下文敏感、基于损失的层、神经元与权重级别迭代优化方法，通过损失差异与阈值对齐决策，并利用回滚机制防止有害更新。该方法克服了Fisher等方法的缺陷，能显著提升合并后网络性能。
原文摘要 · Abstract (English)
Merging neural networks without retraining is central to federated and distributed learning. Common methods such as weight averaging or Fisher merging often lose accuracy and are unstable across seeds. CoGraM (Contextual Granular Merging) is a multi-stage, context-sensitive, loss-based, and iterative optimization method across layers, neurons, and weight levels that aligns decisions with loss differences and thresholds and prevents harmful updates through rollback. CoGraM is an optimization method that addresses the weaknesses of methods such as Fisher and can significantly improve the merged network.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。