English

Online Multi-Agent Decentralized Byzantine-robust Gradient Estimation

Artificial Intelligence 2022-10-03 v1

Abstract

In this paper, we propose an iterative scheme for distributed Byzantineresilient estimation of a gradient associated with a black-box model. Our algorithm is based on simultaneous perturbation, secure state estimation and two-timescale stochastic approximations. We also show the performance of our algorithm through numerical experiments.

Keywords

Cite

@article{arxiv.2209.15274,
  title  = {Online Multi-Agent Decentralized Byzantine-robust Gradient Estimation},
  author = {Alexandre Reiffers-Masson and Isabel Amigo},
  journal= {arXiv preprint arXiv:2209.15274},
  year   = {2022}
}
R2 v1 2026-06-28T02:26:06.655Z