A zeroth-order proximal-gradient method with Gaussian smoothing is proved to find approximate stationary points of stochastic bilevel inverse problems with non-smooth convex lower levels, at oracle complexity O(epsilon^-3) in the nonconvex case and with explicit accuracy in the convex case.
Modern regularizatio n methods for inverse problems
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Derivative-free stochastic bilevel optimization for inverse problems
A zeroth-order proximal-gradient method with Gaussian smoothing is proved to find approximate stationary points of stochastic bilevel inverse problems with non-smooth convex lower levels, at oracle complexity O(epsilon^-3) in the nonconvex case and with explicit accuracy in the convex case.