Journées Julia et Optimisation

Session

Optimisation non-linéaire

6 oct. 2023, 09:00

Jean-Baptiste Say (CNAM Paris)

2. Solving optimal control problems with Julia

Jean-Baptiste Caillau (Université Côte d’Azur, CNRS, Inria, LJAD), Joseph Gergaud (Université de Toulouse, INP-ENSEEIHT-IRIT)

06/10/2023 09:00

J.-B. Caillau, O. Cots, J. Gergaud, P. Martinon, S. Sed The numerical solution of optimal control of dynamical systems is a rich process that typically involves modelling, optimisation, differential equations (most notably Hamiltonian ones), nonlinear equations (e.g. for shooting), pathfollowing methods… and, at every step, automatic differentiation. While tremendous efforts are being made in...

16. Latest developments for mixed-integer nonlinear programming in Artelys Knitro

Maxime Dufour (Artelys)

06/10/2023 10:00

Artelys Knitro is a mathematical programming solver for nonlinear and mixed-integer nonlinear problems. As input, it accepts linear structures, quadratic structures and black-box functions, with if possible, their first and second-order derivatives. Knitro relies on derivative-based algorithms to find locally optimal solutions. Knitro finds the global optimum for convex problems. For...

12. Julia Smooth Optimizers: Streamlined Research and Recent Developments of the Ecosystem

Abel Soares SIQUEIRA (Netherlands eScience Center)

06/10/2023 11:00

Julia Smooth Optimizers (JSO) is an organization created in 2015 focused on streamlining the development of research-level optimization solvers and providing high-performance linear algebra and optimization packages for end-users. We are the maintainers of the packages NLPModels.jl, CUTEst.jl, Krylov.jl, LinearOperators.jl, Percival.jl, JSOSolvers.jl, and around 40 other packages. In this...

27. Convex mixed-integer optimization algorithms with Boscia.jl and applications

Mathieu BESANÇON (Zuse Institute Berlin)

06/10/2023 11:30

Nonlinear and mixed-integer optimization have long remained separate fields with their own techniques, representations, and algorithms.
In this talk, we will introduce a novel approach leveraging first-order methods for convex optimization based on the Frank-Wolfe algorithm.
First-order methods are the usual choices for large-scale smooth optimization but are typically not prime candidates...

30. Synergy between Macaulay and Sum-of-Squares frameworks

Benoit Legat (KU Leuven)

06/10/2023 12:00

Many important optimization problems do not exhibit the convexity feature that would allow them to be solved to global optimality using classical convex optimization methods. In notable cases, such as with neural networks, we even fail to see what structure could be exploited to ensure global optimality. However, a wide variety of smooth nonlinear optimization problems can often be...