FedRLHF: A Convergence-Guaranteed Federated Framework for Privacy-Preserving and Personalized RLHF
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909483574231040 |
|---|---|
| author | Fan, Flint Xiaofeng Tan, Cheston Ong, Yew-Soon Wattenhofer, Roger Ooi, Wei-Tsang |
| author_facet | Fan, Flint Xiaofeng Tan, Cheston Ong, Yew-Soon Wattenhofer, Roger Ooi, Wei-Tsang |
| contents | In the era of increasing privacy concerns and demand for personalized experiences, traditional Reinforcement Learning with Human Feedback (RLHF) frameworks face significant challenges due to their reliance on centralized data. We introduce Federated Reinforcement Learning with Human Feedback (FedRLHF), a novel framework that decentralizes the RLHF process. FedRLHF enables collaborative policy learning across multiple clients without necessitating the sharing of raw data or human feedback, thereby ensuring robust privacy preservation. Leveraging federated reinforcement learning, each client integrates human feedback locally into their reward functions and updates their policies through personalized RLHF processes. We establish rigorous theoretical foundations for FedRLHF, providing convergence guarantees, and deriving sample complexity bounds that scale efficiently with the number of clients. Empirical evaluations on the MovieLens and IMDb datasets demonstrate that FedRLHF not only preserves user privacy but also achieves performance on par with centralized RLHF, while enhancing personalization across diverse client environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_15538 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | FedRLHF: A Convergence-Guaranteed Federated Framework for Privacy-Preserving and Personalized RLHF Fan, Flint Xiaofeng Tan, Cheston Ong, Yew-Soon Wattenhofer, Roger Ooi, Wei-Tsang Machine Learning Artificial Intelligence Cryptography and Security I.2.11 In the era of increasing privacy concerns and demand for personalized experiences, traditional Reinforcement Learning with Human Feedback (RLHF) frameworks face significant challenges due to their reliance on centralized data. We introduce Federated Reinforcement Learning with Human Feedback (FedRLHF), a novel framework that decentralizes the RLHF process. FedRLHF enables collaborative policy learning across multiple clients without necessitating the sharing of raw data or human feedback, thereby ensuring robust privacy preservation. Leveraging federated reinforcement learning, each client integrates human feedback locally into their reward functions and updates their policies through personalized RLHF processes. We establish rigorous theoretical foundations for FedRLHF, providing convergence guarantees, and deriving sample complexity bounds that scale efficiently with the number of clients. Empirical evaluations on the MovieLens and IMDb datasets demonstrate that FedRLHF not only preserves user privacy but also achieves performance on par with centralized RLHF, while enhancing personalization across diverse client environments. |
| title | FedRLHF: A Convergence-Guaranteed Federated Framework for Privacy-Preserving and Personalized RLHF |
| topic | Machine Learning Artificial Intelligence Cryptography and Security I.2.11 |
| url | https://arxiv.org/abs/2412.15538 |