Online Multi-Agent Decentralized Byzantine-robust Gradient Estimation
Résumé
In this paper, we propose an iterative scheme for distributed Byzantine-resilient 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.
Domaines
Informatique [cs]| Origine | Fichiers produits par l'(les) auteur(s) |
|---|