ZO-MGT achieves O(1/T) convergence in distributed zeroth-order optimization while suppressing heterogeneity bias at quadratic rate O((1-β)^2) using momentum and Rademacher perturbations.
Distributed zeroth-order gradi- ent tracking for weakly convex optimization over unbalanced graphs
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Distributed Zeroth-Order Optimization with Rademacher Perturbations and Momentum Gradient Tracking
ZO-MGT achieves O(1/T) convergence in distributed zeroth-order optimization while suppressing heterogeneity bias at quadratic rate O((1-β)^2) using momentum and Rademacher perturbations.