Learning to Learn Weight Generation via Local Consistency Diffusion

TL;DR

This paper introduces Mc-Di, a diffusion-based meta-learning approach that improves weight generation by addressing local-global consistency and transferability, enhancing performance in various learning tasks.

Contribution

It integrates diffusion algorithms with meta-learning and extends diffusion to local consistency, enabling better generalization and local target learning.

Findings

Outperforms existing methods in transfer and few-shot learning tasks.

Demonstrates improved accuracy and inference efficiency.

Maintains consistency with global optima while learning from local targets.

Abstract

Diffusion-based algorithms have emerged as promising techniques for weight generation. However, existing solutions are limited by two challenges: generalizability and local target assignment. The former arises from the inherent lack of cross-task transferability in existing single-level optimization methods, limiting the model's performance on new tasks. The latter lies in existing research modeling only global optimal weights, neglecting the supervision signals in local target weights. Moreover, naively assigning local target weights causes local-global inconsistency. To address these issues, we propose Mc-Di, which integrates the diffusion algorithm with meta-learning for better generalizability. Furthermore, we extend the vanilla diffusion into a local consistency diffusion algorithm. Our theory and experiments demonstrate that it can learn from local targets while maintaining consistency with the global optima. We validate Mc-Di's superior accuracy and inference efficiency in tasks that require frequent weight updates, including transfer learning, few-shot learning, domain generalization, and large language model adaptation.