Point-Based Set-Valued Approximations, C -Differential Operators and Applications
L. Qi introduced the \(C\)-differential operator in 1996 [\(C\)-Differential operators, \(C\)-differentiability and generalized Newton methods, Applied Mathematics Report, AMR96/5, School of Mathematics, The University of New South Wales, Sydney, Australia] that since then has attracted a considerable attention in the nonsmooth optimization area. However, a number of theoretical questions regarding \(C\)-differential operators have not been answered. In particular, the four questions posed by L. Qi in 1996. The paper presents a theoretical investigation on point-base set-valued approximation and the \(C\)-differential operator. This paper tries to answer some of the questions posed by L. Qi in 1996 in an anticipation of getting a better understanding of the \(C\)-differential operator and its relationship with point-based set-valued approximation. This paper proves that a continuous vector valued function \(F\) from \(\mathbb R^n\) to \(\mathbb R^n\) is locally Lipschitz at a point if and only if \(F\) admits a \(C\)-differential operator in a neighbourhood of the point. This paper proves that if \(F\) is Fréchet differentiable at a point \(x\), then the Jacobian \(\nabla F(x)\) at \(x\) is contained in the plenary hull of a \(C\)-differential operator of \(F\) does not necessary coincide with the Clarke generalized Jacobian \(\partial F\) even if \(F\) is locally Lipschitz. The author applies the notion of \(C\)-differential operator to nonsmooth minimization. This paper derives optimality conditions and mean valued theorems in terms of convex \(C\)-differential operators and proposes a conceptual algorithm using \(C\)-differential operators. The results of this paper represent a contribution to nonsmooth optimization.
- A minimal set-valued strong derivative for vector-valued Lipschitz functions
- A nonsmooth version of Newton's method
- Adaptive smoothing method, deterministically computable generalized Jacobians, and the Newton method
- Approximate Jacobian Matrices for Nonsmooth Continuous Maps and C1-Optimization
- Approximate Newton methods for nonsmooth equations
- Inexact generalized Newton methods for second order C-differentiable optimization
- Newton's method for a class of nonsmooth functions
- Nonlinear programming and nonsmooth optimization by successive linear programming
- Nonsmooth Equations: Motivation and Algorithms
- On the Extension of Constrained Optimization Algorithms from Differentiable to Nondifferentiable Problems
- Optimization and nonsmooth analysis
- Set-valued approximations and Newton's methods
- Smoothing Methods and Semismooth Methods for Nondifferentiable Operator Equations
This page was built for publication: Point-Based Set-Valued Approximations, C -Differential Operators and Applications
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4430665)