Characterization of nonsmooth functions through their generalized gradients
- 1 January 1991
- journal article
- research article
- Published by Taylor & Francis in Optimization
- Vol. 22 (3), 401-416
- https://doi.org/10.1080/02331939108843678
Abstract
Given a real-valued Lipschitz function, we explore properties of its generalized gradient in Clarke's sense that corresponds to f being quasi-convex, to f being d.c, (difference of convex functions), to f being a pointwise supremum of functions that are ktimes continuously differentiable. In other words, knowing that ∂f is the generalized gradient multifunction of a function f what kind of properties of ∂f could serve to characterize f?. The classes of nonsmooth function involed in this paper are: quasi-function, d.e functions, lower-C k functions, semi smooth functions and quasidifferentiable functions.Keywords
This publication has 16 references indexed in Scilit:
- The conjugate of the difference of convex functionsJournal of Optimization Theory and Applications, 1986
- On quasidifferentiable mappingsMathematische Operationsforschung und Statistik. Series Optimization, 1983
- Generalized gradients of Lipschitz functionalsAdvances in Mathematics, 1981
- Submonotone subdifferentials of Lipschitz functionsTransactions of the American Mathematical Society, 1981
- Minimization of a quasi-differentiable function in a quasi-differentiable setUSSR Computational Mathematics and Mathematical Physics, 1980
- Generic differentiability of Lipschitzian functionsTransactions of the American Mathematical Society, 1979
- Calcul sous-differentiel et optimisationJournal of Functional Analysis, 1978
- Semismooth and Semiconvex Functions in Constrained OptimizationSIAM Journal on Control and Optimization, 1977
- Generalized gradients and applicationsTransactions of the American Mathematical Society, 1975
- Convex AnalysisPublished by Walter de Gruyter GmbH ,1970