Paper ID: 2209.15274

Online Multi-Agent Decentralized Byzantine-robust Gradient Estimation

Alexandre Reiffers-Masson, Isabel Amigo

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.

Submitted: Sep 30, 2022