EF-Landing provably converges at O(1/sqrt(N K)) for distributed stochastic problems on the Stiefel manifold while using compressed communication and no retraction.
Infeasible Deterministic , Stochastic , and Variance - Reduction Algorithms for Optimization under Orthogonality Constraints
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Distributed Retraction-Free and Communication-Efficient Optimization on the Stiefel Manifold
EF-Landing provably converges at O(1/sqrt(N K)) for distributed stochastic problems on the Stiefel manifold while using compressed communication and no retraction.