| Paper ID | SPTM-10.6 |
| Paper Title |
Graph-Homomorphic Perturbations for Private Decentralized Learning |
| Authors |
Stefan Vlaski, Ali H. Sayed, Ecole Polytechnique Fédérale de Lausanne (EPFL), Switzerland |
| Session | SPTM-10: Distributed Learning over Graphs |
| Location | Gather.Town |
| Session Time: | Wednesday, 09 June, 14:00 - 14:45 |
| Presentation Time: | Wednesday, 09 June, 14:00 - 14:45 |
| Presentation |
Poster
|
| Topic |
Signal Processing Theory and Methods: [SIPG] Signal and Information Processing over Graphs |
| IEEE Xplore Open Preview |
Click here to view in IEEE Xplore |
| Virtual Presentation |
Click here to watch in the Virtual Conference |
| Abstract |
Decentralized algorithms for stochastic optimization and learning rely on the diffusion of information through repeated local exchanges of intermediate estimates. Such structures are particularly appealing in situations where agents may be hesitant to share raw data due to privacy concerns. Nevertheless, in the absence of additional privacy-preserving mechanisms, the exchange of local estimates, which are generated based on private data can allow for the inference of the data itself. The most common mechanism for guaranteeing privacy is the addition of perturbations to local estimates before broadcasting. These perturbations are generally chosen independently at every agent, resulting in a significant performance loss. We propose an alternative scheme, which constructs perturbations according to a particular nullspace condition, allowing them to be invisible (to first order in the step-size) to the network centroid, while preserving privacy guarantees. The analysis allows for general nonconvex loss functions, and is hence applicable to a large number of machine learning and signal processing problems, including deep learning. |
