A SPIDER-type stochastic subgradient method with smoothed exact penalization reaches (epsilon,epsilon)-KKT points of expectation-constrained nonconvex nonsmooth problems in O(epsilon^-4) iterations.
Title resolution pending
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
-
A single-loop SPIDER-type stochastic subgradient method for expectation-constrained nonconvex nonsmooth optimization
A SPIDER-type stochastic subgradient method with smoothed exact penalization reaches (epsilon,epsilon)-KKT points of expectation-constrained nonconvex nonsmooth problems in O(epsilon^-4) iterations.