Proposes DCSMD-SW for distributed composite optimization over time-varying networks, deriving high-probability convergence rates under sub-Weibull gradient noise without smoothness assumptions.
Optimization for data-driven learning and control
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
High-Probability Convergence Theory for Distributed Composite Optimization with Sub-Weibull Noises
Proposes DCSMD-SW for distributed composite optimization over time-varying networks, deriving high-probability convergence rates under sub-Weibull gradient noise without smoothness assumptions.