Compositions of convex functions and fully linear models
From MaRDI portal
Abstract: Derivative-free optimization (DFO) is the mathematical study of the optimization algorithms that do not use derivatives. One branch of DFO focuses on model-based DFO methods, where an approximation of the objective function is used to guide the optimization algorithm. Proving convergence of such methods often applies an assumption that the approximations form {em fully linear models} -- an assumption that requires the true objective function to be smooth. However, some recent methods have loosened this assumption and instead worked with functions that are compositions of smooth functions with simple convex functions (the max-function or the norm). In this paper, we examine the error bounds resulting from the composition of a convex lower semi-continuous function with a smooth vector-valued function when it is possible to provide fully linear models for each component of the vector-valued function. We derive error bounds for the resulting function values and subgradient vectors.
Recommendations
- A derivative-free comirror algorithm for convex optimization
- Derivative-free optimization via proximal point methods
- A derivative-free trust-region algorithm for composite nonsmooth optimization
- A derivative-free \(\mathcal{V} \mathcal{U}\)-algorithm for convex finite-max problems
- Model-based derivative-free methods for convex-constrained optimization
Cites work
- A derivative-free approximate gradient sampling algorithm for finite minimax problems
- A derivative-free comirror algorithm for convex optimization
- Convex analysis and monotone operator theory in Hilbert spaces
- Derivative-free optimization via proximal point methods
- Developments of NEWUOA for minimization without derivatives
- Geometry of interpolation sets in derivative free optimization
- Geometry of sample sets in derivative-free optimization: polynomial regression and underdetermined interpolation
- Global convergence of general derivative-free trust-region algorithms to first- and second-order critical points
- Global convergence of radial basis function trust region derivative-free algorithms
- scientific article; zbMATH DE number 1971709 (Why is no real title available?)
- Introduction to Derivative-Free Optimization
- Manifold sampling for _1 nonconvex optimization
- On trust region methods for unconstrained minimization without derivatives
- Optimizing damper connectors for adjacent buildings
- The calculus of simplex gradients
- UOBYQA: unconstrained optimization by quadratic approximation
- Using Sampling and Simplex Derivatives in Pattern Search Methods
- Variational Analysis
Cited in
(7)- Derivative-free robust optimization by outer approximations
- Manifold sampling for optimization of nonconvex functions that are piecewise linear compositions of smooth components
- Derivative-free optimization methods
- Calculus identities for generalized simplex gradients: rules and applications
- Error Analysis of Surrogate Models Constructed Through Operations on Submodels
- \(Q\)-fully quadratic modeling and its application in a random subspace derivative-free method
- A discussion on variational analysis in derivative-free optimization
This page was built for publication: Compositions of convex functions and fully linear models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1679619)
