On the Privacy-Robustness-Utility Trilemma in Distributed Learning
The ubiquity of distributed machine learning (ML) in sensitive public domain applications calls for algorithms that protect data privacy, while being robust to faults and adversarial behaviors. Although privacy and robustness have been extensively studied independently in distributed ML, their synth...
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Artikel |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2302.04787 http://arxiv.org/abs/2302.04787 |
| Tags: |
Keine Tags, Fügen Sie das erste Tag hinzu!
|