Secure Federated Learning Across Heterogeneous Cloud and High-Performance Computing Resources -- A Case Study on Federated Fine-tuning of LLaMA 2

arXiv:2402.12271v1 Announce Type: cross
Abstract: Federated learning enables multiple data owners to collaboratively train robust machine learning models without transferring large or sensitive local datasets by only sharing the parameters of the locally trained models. In this paper, we elaborate on the design of our Advanced Privacy-Preserving Federated Learning (APPFL) framework, which streamlines end-to-end secure and reliable federated learning experiments across cloud computing facilities and high-performance computing resources by leveraging Globus Compute, a distributed function as a service platform, and …

abstract arxiv case case study cloud computing computing resources cs.dc cs.lg data datasets federated learning fine-tuning llama llama 2 machine machine learning machine learning models multiple paper parameters performance resources robust study train type