Improving the communication in decentralized manifold optimization through single-step consensus and compression

Documenti analoghi

• Riemannian EXTRA: Communication-efficient decentralized optimization over compact submanifolds with data heterogeneity
  di: Wu, Jiayuan, et al.
  Pubblicazione: (2025)
• An Augmented Lagrangian Primal-Dual Semismooth Newton Method for Multi-Block Composite Optimization
  di: Deng, Zhanwang, et al.
  Pubblicazione: (2023)
• An efficient primal dual semismooth Newton method for semidefinite programming
  di: Deng, Zhanwang, et al.
  Pubblicazione: (2025)
• Single-loop $\mathcal{O}(ε^{-3})$ stochastic smoothing algorithms for nonsmooth Riemannian optimization
  di: Deng, Kangkang, et al.
  Pubblicazione: (2025)
• Adaptive Riemannian ADMM for Nonsmooth Optimization: Optimal Complexity without Smoothing
  di: Deng, Kangkang, et al.
  Pubblicazione: (2025)