Revisiting Gradient Pruning: A Dual Realization for Defending against Gradient Attacks
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914659022405632 |
|---|---|
| author | Xue, Lulu Hu, Shengshan Zhao, Ruizhi Zhang, Leo Yu Hu, Shengqing Sun, Lichao Yao, Dezhong |
| author_facet | Xue, Lulu Hu, Shengshan Zhao, Ruizhi Zhang, Leo Yu Hu, Shengqing Sun, Lichao Yao, Dezhong |
| contents | Collaborative learning (CL) is a distributed learning framework that aims to protect user privacy by allowing users to jointly train a model by sharing their gradient updates only. However, gradient inversion attacks (GIAs), which recover users' training data from shared gradients, impose severe privacy threats to CL. Existing defense methods adopt different techniques, e.g., differential privacy, cryptography, and perturbation defenses, to defend against the GIAs. Nevertheless, all current defense methods suffer from a poor trade-off between privacy, utility, and efficiency. To mitigate the weaknesses of existing solutions, we propose a novel defense method, Dual Gradient Pruning (DGP), based on gradient pruning, which can improve communication efficiency while preserving the utility and privacy of CL. Specifically, DGP slightly changes gradient pruning with a stronger privacy guarantee. And DGP can also significantly improve communication efficiency with a theoretical analysis of its convergence and generalization. Our extensive experiments show that DGP can effectively defend against the most powerful GIAs and reduce the communication cost without sacrificing the model's utility. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_16687 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Revisiting Gradient Pruning: A Dual Realization for Defending against Gradient Attacks Xue, Lulu Hu, Shengshan Zhao, Ruizhi Zhang, Leo Yu Hu, Shengqing Sun, Lichao Yao, Dezhong Cryptography and Security Machine Learning Collaborative learning (CL) is a distributed learning framework that aims to protect user privacy by allowing users to jointly train a model by sharing their gradient updates only. However, gradient inversion attacks (GIAs), which recover users' training data from shared gradients, impose severe privacy threats to CL. Existing defense methods adopt different techniques, e.g., differential privacy, cryptography, and perturbation defenses, to defend against the GIAs. Nevertheless, all current defense methods suffer from a poor trade-off between privacy, utility, and efficiency. To mitigate the weaknesses of existing solutions, we propose a novel defense method, Dual Gradient Pruning (DGP), based on gradient pruning, which can improve communication efficiency while preserving the utility and privacy of CL. Specifically, DGP slightly changes gradient pruning with a stronger privacy guarantee. And DGP can also significantly improve communication efficiency with a theoretical analysis of its convergence and generalization. Our extensive experiments show that DGP can effectively defend against the most powerful GIAs and reduce the communication cost without sacrificing the model's utility. |
| title | Revisiting Gradient Pruning: A Dual Realization for Defending against Gradient Attacks |
| topic | Cryptography and Security Machine Learning |
| url | https://arxiv.org/abs/2401.16687 |