Derivative-free optimization methods
In many optimization problems arising from scientific, engineering and artificial intelligence
applications, objective and constraint functions are available only as the output of a black …
applications, objective and constraint functions are available only as the output of a black …
Conservative set valued fields, automatic differentiation, stochastic gradient methods and deep learning
Modern problems in AI or in numerical analysis require nonsmooth approaches with a
flexible calculus. We introduce generalized derivatives called conservative fields for which …
flexible calculus. We introduce generalized derivatives called conservative fields for which …
A mathematical model for automatic differentiation in machine learning
Automatic differentiation, as implemented today, does not have a simple mathematical
model adapted to the needs of modern machine learning. In this work we articulate the …
model adapted to the needs of modern machine learning. In this work we articulate the …
On correctness of automatic differentiation for non-differentiable functions
Differentiation lies at the core of many machine-learning algorithms, and is well-supported
by popular autodiff systems, such as TensorFlow and PyTorch. Originally, these systems …
by popular autodiff systems, such as TensorFlow and PyTorch. Originally, these systems …
Computationally relevant generalized derivatives: theory, evaluation and applications
A new method for evaluating generalized derivatives in nonsmooth problems is reviewed.
Lexicographic directional (LD-) derivatives are a recently developed tool in nonsmooth …
Lexicographic directional (LD-) derivatives are a recently developed tool in nonsmooth …
First-and second-order optimality conditions for piecewise smooth objective functions
A Griewank, A Walther - Optimization Methods and Software, 2016 - Taylor & Francis
Any piecewise smooth function that is specified by an evaluation procedure involving
smooth elemental functions and piecewise linear functions like and can be represented in …
smooth elemental functions and piecewise linear functions like and can be represented in …
Relaxing kink qualifications and proving convergence rates in piecewise smooth optimization
A Griewank, A Walther - SIAM Journal on Optimization, 2019 - SIAM
In the paper [Optim. Methods Softw., 31 (2016), pp. 904--930] we derived first order (KKT)
and second order (second order sufficiency condition (SOSC)) optimality conditions for …
and second order (second order sufficiency condition (SOSC)) optimality conditions for …
Branch-locking AD techniques for nonsmooth composite functions and nonsmooth implicit functions
KA Khan - Optimization Methods and Software, 2018 - Taylor & Francis
A recent nonsmooth vector forward mode of algorithmic differentiation (AD) computes
Nesterov's L-derivatives for nonsmooth composite functions; these L-derivatives provide …
Nesterov's L-derivatives for nonsmooth composite functions; these L-derivatives provide …
Manifold Sampling for Nonconvex Optimization
We present a new algorithm, called manifold sampling, for the unconstrained minimization of
a nonsmooth composite function h∘F when h has known structure. In particular, by …
a nonsmooth composite function h∘F when h has known structure. In particular, by …
An algorithm for nonsmooth optimization by successive piecewise linearization
S Fiege, A Walther, A Griewank - Mathematical Programming, 2019 - Springer
We present an optimization method for Lipschitz continuous, piecewise smooth (PS)
objective functions based on successive piecewise linearization. Since, in many realistic …
objective functions based on successive piecewise linearization. Since, in many realistic …