LOGSAFE: Logic-Guided Verification for Trustworthy Federated Time-Series Learning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908907459313664 |
|---|---|
| author | Nguyen, Dung Thuy An, Ziyan Johnson, Taylor T. Ma, Meiyi Leach, Kevin |
| author_facet | Nguyen, Dung Thuy An, Ziyan Johnson, Taylor T. Ma, Meiyi Leach, Kevin |
| contents | This paper introduces LOGSAFE, a defense mechanism for federated learning in time series settings, particularly within cyber-physical systems. It addresses poisoning attacks by moving beyond traditional update-similarity methods and instead using logical reasoning to evaluate client reliability. LOGSAFE extracts client-specific temporal properties, infers global patterns, and verifies clients against them to detect and exclude malicious participants. Experiments show that it significantly outperforms existing methods, achieving up to 93.27% error reduction over the next best baseline. Our code is available at https://github.com/judydnguyen/LOGSAFE-Robust-FTS. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_03231 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | LOGSAFE: Logic-Guided Verification for Trustworthy Federated Time-Series Learning Nguyen, Dung Thuy An, Ziyan Johnson, Taylor T. Ma, Meiyi Leach, Kevin Cryptography and Security Artificial Intelligence Distributed, Parallel, and Cluster Computing Logic in Computer Science This paper introduces LOGSAFE, a defense mechanism for federated learning in time series settings, particularly within cyber-physical systems. It addresses poisoning attacks by moving beyond traditional update-similarity methods and instead using logical reasoning to evaluate client reliability. LOGSAFE extracts client-specific temporal properties, infers global patterns, and verifies clients against them to detect and exclude malicious participants. Experiments show that it significantly outperforms existing methods, achieving up to 93.27% error reduction over the next best baseline. Our code is available at https://github.com/judydnguyen/LOGSAFE-Robust-FTS. |
| title | LOGSAFE: Logic-Guided Verification for Trustworthy Federated Time-Series Learning |
| topic | Cryptography and Security Artificial Intelligence Distributed, Parallel, and Cluster Computing Logic in Computer Science |
| url | https://arxiv.org/abs/2411.03231 |