Federated Learning-Based Privacy-Preserving Fine-tuning of Large Language Models for Personalized Generative Applications
Keywords:
federated learning, large language models, privacy-preserving, fine-tuning, decentralized trainingAbstract
Fine-tuning of large language models (LLMs) with the help of transformative approach of Federated learning (FL) while preserving user privacy in personalized generative applications. The objective of the research paper is to introduce a privacy-preserving FL-based framework that can enable decentralized fine-tuning of LLMs for distributed user-specific datasets without exposing individual data.
Downloads
References
M. McMahan, E. Moore, D. Ramage, S. H. J. Arcas, and B. y. P. H. Zhao, “Communication-efficient learning of deep networks from decentralized data,” Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS), 2017, pp. 1273–1282.
H. Shokri and V. Shmatikov, “Privacy-preserving deep learning,” Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security (CCS), 2015, pp. 1310–1321.
S. R. K. Gade, S. R. M. Parikh, and A. R. Soni, “A survey on federated learning: Its applications, challenges, and opportunities,” Journal of King Saud University - Computer and Information Sciences, vol. 34, no. 8, pp. 6226–6237, 2022.
F. Yang, Z. Liu, S. Zhang, Z. Li, and Y. Liu, “Federated learning for mobile edge networks: A comprehensive survey,” IEEE Transactions on Industrial Informatics, vol. 18, no. 4, pp. 2293–2307, Apr. 2022.
A. Choudhury and N. M. Amato, “Privacy-preserving federated learning: Challenges, methods, and future directions,” IEEE Transactions on Knowledge and Data Engineering, vol. 34, no. 1, pp. 97–109, Jan. 2022.
L. Zhang, J. Chen, and Q. Li, “Differential privacy in machine learning: A comprehensive review,” IEEE Transactions on Knowledge and Data Engineering, vol. 34, no. 3, pp. 1804–1816, Mar. 2022.
X. Xu, T. Li, W. Lou, and H. Zeng, “Secure aggregation for federated learning: A comprehensive survey,” IEEE Access, vol. 9, pp. 32902–32915, 2021.
M. Bonawitz, H. Eichner, W. Grieskamp, et al., “Towards federated learning at scale: System design,” Proceedings of the 2nd Conference on Machine Learning and Systems, 2020, pp. 374–388.
C. Geyer, S. Klein, and S. K. M. Sundaram, “Differentially private federated learning: A survey,” IEEE Transactions on Neural Networks and Learning Systems, vol. 33, no. 6, pp. 2677–2691, Jun. 2022.
D. Kairouz, H. B. McMahan, B. N. V. S. Papalambros, et al., “Advances and open problems in federated learning,” Foundations and Trends® in Machine Learning, vol. 14, no. 1, pp. 1–210, 2021.
X. Xie, Z. Wang, and J. Yang, “A review on the security of federated learning,” IEEE Access, vol. 8, pp. 147423–147439, 2020.
A. Y. Tong, L. Yang, and K. A. Hsieh, “Federated learning with privacy-preserving differential privacy,” Proceedings of the 23rd International Conference on Machine Learning (ICML), 2021, pp. 1210–1220.
A. D. Joseph, E. R. M. Raghu, and R. S. Gupta, “Security and privacy in federated learning: Challenges and solutions,” IEEE Transactions on Computational Social Systems, vol. 9, no. 6, pp. 1327–1338, Dec. 2022.
G. Zyskind, O. Nathan, and A. Pentland, “Decentralizing privacy: Using blockchain to protect personal data,” Proceedings of the IEEE Security and Privacy Workshops, 2015, pp. 180–184.
R. Shokri, S. V. S. Shmatikov, and M. Springer, “Privacy-preserving deep learning,” Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security (CCS), 2015, pp. 1–9.
J. Lyu, X. Zhang, and D. S. T. Tsai, “Towards secure federated learning: A comprehensive survey,” IEEE Transactions on Network and Service Management, vol. 19, no. 2, pp. 1229–1245, Jun. 2022.
M. Nasr, F. Wang, and D. P. Wood, “Comprehensive privacy in federated learning,” Proceedings of the 2021 IEEE Symposium on Security and Privacy (SP), 2021, pp. 2062–2079.
Y. Wu, X. Zhang, and M. Y. I. Li, “Secure federated learning with differential privacy for privacy-preserving artificial intelligence,” IEEE Transactions on Artificial Intelligence, vol. 3, no. 5, pp. 321–334, Oct. 2022.
P. L. Bartlett, F. M. B. Müller, and H. B. McMahan, “Model-agnostic privacy-preserving federated learning with differential privacy,” Proceedings of the 35th International Conference on Machine Learning (ICML), 2018, pp. 3861–3870.
S. R. K. Gade, R. M. Soni, and A. K. Parikh, “Federated learning and its applications in artificial intelligence: A survey and future prospects,” Journal of Cloud Computing: Advances, Systems and Applications, vol. 11, no. 1, pp. 1–17, Dec. 2022.
