EUVIMEDEUVIMED entdecken
Crossref

Byzantine-Resilient Decentralized Optimization for Joint Feature Selection in Multi-Task Networks

Dazhong Wang, Weidong Liu, Xiaojun Mao

IEEE Transactions on Signal Processing · 2026 · 74 · 1926-1941

DOI: 10.1109/tsp.2026.3688176 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Signal Processing, 2026. Verfasst von Dazhong Wang und 2 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper initiates a study of joint feature selection in multi-task networks, where agents collaboratively learn related prediction models over decentralized networks while keeping data local. Despite extensive progress in decentralized optimization and Byzantine-robust consensus, their integration to realize fully decentralized $\boldsymbol{\ell}_{\mathbf{2,1}}$ -coupled multi-task feature learning remains an open challenge. To address this gap, the paper proposes a Byzantine-resilient decentralized proximal gradient framework, named Brd-MTFL, for joint feature selection and model learning under the $\boldsymbol{\ell}_{\mathbf{2,1}}$ -regularization. The core innovation lies in a fully decentralized row-norm consensus mechanism that allows agents to compute global $\boldsymbol{\ell}_{\mathbf 2}$ row norms associated with feature groups through peer-to-peer communication, thus realizing fully decentralized joint feature selection. To counter malicious or faulty updates, a trimmed-mean-based robust aggregation rule is integrated into the row-norm estimation step, providing provable robustness to Byzantine failures. Our theoretical analysis establishes that Brd-MTFL achieves near-optimal convergence rates, thereby closing a critical theoretical gap at the intersection of Byzantine resilience, decentralized consensus on intermediate quantities, and $\boldsymbol{\ell}_{\mathbf{2,1}}$ -coupled feature selection in multi-task networks. Specifically, we demonstrate finite-time $\boldsymbol{O}\mathbf{(1/}\boldsymbol{T}\mathbf{)}$ convergence under convexity and linear convergence under strong convexity, with both guarantees explicitly characterized by their dependence on network topology and the Byzantine budget. Empirical evaluations on synthetic data and real data demonstrate that Brd-MTFL achieves superior accuracy, stability, and resilience.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Wang, D., Liu, W., & Mao, X. (2026). Byzantine-Resilient Decentralized Optimization for Joint Feature Selection in Multi-Task Networks. IEEE Transactions on Signal Processing, 74, 1926-1941. https://doi.org/10.1109/tsp.2026.3688176

Vancouver

Wang D, Liu W, Mao X. Byzantine-Resilient Decentralized Optimization for Joint Feature Selection in Multi-Task Networks. IEEE Transactions on Signal Processing. 2026;74:1926-1941. doi:10.1109/tsp.2026.3688176

BibTeX
@article{Wang2026,
  author = {Wang, Dazhong and Liu, Weidong and Mao, Xiaojun},
  title = {Byzantine-Resilient Decentralized Optimization for Joint Feature Selection in Multi-Task Networks},
  journal = {IEEE Transactions on Signal Processing},
  year = {2026},
  volume = {74},
  pages = {1926--1941},
  doi = {10.1109/tsp.2026.3688176}
}
Kostenlos starten →