A Chain Rule for Essentially Smooth Lipschitz Functions
From MaRDI portal
A concept of arcwise essentially smooth functions is defined. This is a generalization of the earlier concept of essentially smooth functions which says that the set of points where \[ f^\circ(x;y) \not= - f^\circ(x;-y), \] (\(f^\circ(x;y)\) is the Clarke generalized derivative) is Haar-null. It is shown that the class of arcwise essentially smooth functions is a subset of the class of essentially smooth functions. Subsequently a chain rule is established for arcwise essentially smooth functions.
Recommendations
Cited in
(16)- Chain rules for nonsmooth functions
- Essentially smooth Lipschitz functions
- Links between functions and subdifferentials
- Upper semismooth functions and the subdifferential determination property
- Conservative set valued fields, automatic differentiation, stochastic gradient methods and deep learning
- Generalized Jacobian for functions with infinite dimensional range and domain
- Nonsmooth bundle trust-region algorithm with applications to robust stability
- Path differentiability of ODE flows
- Directional upper derivatives and the chain rule formula for locally Lipschitz functions on Banach spaces
- Chain rule for conic derivatives
- A chain rule for nonsmooth composite functions via minimisation
- Essentially smooth Lipschitz functions: compositions and chain rules
- scientific article; zbMATH DE number 1786286 (Why is no real title available?)
- Examples of Pathological Dynamics of the Subgradient Method for Lipschitz Path-Differentiable Functions
- Conservative parametric optimality and the ridge method for tame min-max problems
- Long term dynamics of the subgradient method for Lipschitz path differentiable functions
This page was built for publication: A Chain Rule for Essentially Smooth Lipschitz Functions
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4389193)