FedHFT: Efficient Federated Finetuning with Heterogeneous Edge Clients
This addresses the problem of personalized natural language understanding for edge clients with privacy and resource constraints, representing an incremental advance in federated learning methods.
The paper tackles the challenges of fine-tuning large language models with limited, heterogeneous data and varying client resources by proposing FedHFT, a federated fine-tuning framework that uses masked adapters and bi-level optimization, achieving significant performance and efficiency improvements over existing methods.
Fine-tuning pre-trained large language models (LLMs) has become a common practice for personalized natural language understanding (NLU) applications on downstream tasks and domain-specific datasets. However, there are two main challenges: (i) limited and/or heterogeneous data for fine-tuning due to proprietary data confidentiality or privacy requirements, and (ii) varying computation resources available across participating clients such as edge devices. This paper presents FedHFT - an efficient and personalized federated fine-tuning framework to address both challenges. First, we introduce a mixture of masked adapters to handle resource heterogeneity across participating clients, enabling high-performance collaborative fine-tuning of pre-trained language model(s) across multiple clients in a distributed setting, while keeping proprietary data local. Second, we introduce a bi-level optimization approach to handle non-iid data distribution based on masked personalization and client clustering. Extensive experiments demonstrate significant performance and efficiency improvements over various natural language understanding tasks under data and resource heterogeneity compared to representative heterogeneous federated learning methods.