通过局部一致性扩散学习学习权重生成
Learning to Learn Weight Generation via Local Consistency Diffusion
把扩散模型用于直接生成网络权重，并用局部一致性约束把局部目标权重的监督信号纳入训练，为视觉模型在迁移、少样本、域泛化场景下的快速权重更新提供了一条可复用的元学习式生成思路。
原文依据 we propose Mc-Di, which integrates the diffusion algorithm with meta-learning for better generalizability
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论基于扩散的算法已成为权重生成的有前景技术。然而，现有方案受限于两个挑战：泛化性与缺失的局部监督目标。第一个挑战源于现有单层优化方法本身缺乏跨任务可迁移性，这限制了模型在新任务上的性能。后一个挑战在于现有研究仅对全局最优权重建模，忽略了局部目标权重中的监督信号。此外，朴素地分配局部目标权重会导致局部目标与全局目标之间的不一致。为解决这些问题，我们提出 Mc-Di，它将扩散算法与元学习相结合以获得更好的泛化性。此外，我们将原始扩散扩展为局部一致性扩散算法。我们的理论分析与实验结果表明，该模型能够从局部目标中学习，同时保持与全局最优的一致性。我们在需要频繁更新权重的任务上验证了 Mc-Di 更优的准确性与推理效率，包括迁移学习、少样本学习、域泛化和语言模型微调。
英文原文摘要 1226 字符
Diffusion-based algorithms have emerged as promising techniques for weight generation. However, existing solutions are limited by two challenges: generalizability and missing local supervision targets. The first challenge stems from the inherent lack of cross-task transferability in existing single-level optimization methods, which limits model performance on new tasks. The latter challenge lies in existing research modeling only global optimal weights, neglecting the supervision signals in local target weights. Furthermore, naively assigning local target weights leads to inconsistency between local and global objectives. To address these issues, we propose Mc-Di, which integrates the diffusion algorithm with meta-learning for better generalizability. Additionally, we extend the vanilla diffusion into a local consistency diffusion algorithm. Our theoretical analysis and experimental results demonstrate that the model can learn from local targets while preserving consistency with the global optimum. We validate Mc-Di's superior accuracy and inference efficiency on tasks that require frequent weight updates, including transfer learning, few-shot learning, domain generalization, and language model fine-tuning.