Harmonizing Generalization and Personalization in Federated Prompt Learning

TL;DR

This paper introduces FedPGP, a federated prompt learning method that balances the generalization ability of large vision-language models with client-specific personalization, improving performance across diverse datasets.

Contribution

FedPGP is a novel approach that combines knowledge-guided global prompts with low-rank personalized adaptation to harmonize generalization and personalization in federated learning.

Findings

FedPGP outperforms existing methods in generalization to novel categories and domains.

It effectively balances personalization and generalization across heterogeneous datasets.

Experimental results demonstrate superior performance of FedPGP in various federated learning scenarios.

Abstract

Federated Prompt Learning (FPL) incorporates large pre-trained Vision-Language models (VLM) into federated learning through prompt tuning. The transferable representations and remarkable generalization capacity of VLM make them highly compatible with the integration of federated learning. Addressing data heterogeneity in federated learning requires personalization, but excessive focus on it across clients could compromise the model's ability to generalize effectively. To preserve the impressive generalization capability of VLM, it is crucial to strike a balance between personalization and generalization in FPL. To tackle this challenge, we proposed Federated Prompt Learning with CLIP Generalization and low-rank Personalization (FedPGP), which employs pre-trained CLIP to provide knowledge-guidance on the global prompt for improved generalization and incorporates a low-rank adaptation term to personalize the global prompt. Further, FedPGP integrates a prompt-wise contrastive loss to achieve knowledge guidance and personalized adaptation simultaneously, enabling a harmonious balance between personalization and generalization in FPL. We conduct extensive experiments on various datasets to explore base-to-novel generalization in both category-level and domain-level scenarios with heterogeneous data, showing the superiority of FedPGP in balancing generalization and personalization.