P2PRISM - Peer to peer learning with individual prism for secure aggregation

Atul Sharma, Wei Chen, Joshua Christian Zhao, Qiang Qiu, Saurabh Bagchi, Somali Chaterji

Keywords: peer-to-peer, decentralized learning, byzantine-robust

TL;DR: We highlight the vulnerabilities in peer-to-peer learning towards malicious attacks and propose a byzantine-robust defense against the same.

Abstract: Federated learning (FL) has made collaboration between nodes possible without explicit sharing of local data. However, it requires the participating nodes to trust the server and its model updates, the server itself being a critical node susceptible to failure and compromise. A loss of trust in the server and a demand to aggregate the model independently for oneself has led decentralized peer-to-peer learning (P2PL) to gain traction lately. In this paper, we highlight the never before exposed vulnerabilities of P2PL towards malicious attacks and how P2PL behaves differently from FL in such a malicious environment. We then present a robust defense - P2PRISM as a secure aggregation protocol for P2PL.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: General Machine Learning (ie none of the above)