Workshop Optimization: Fundamentals and Algorithms for Structured Problems

Workshop Optimization: Fundamentals and

Algorithms for Structured Problems

CIMI Workshop, Toulouse, France

28-29th june 2018


Optimization is booming at both the industrial and academic levels, stimulated by the new challenges posed by strategic applications with high societal impact. This workshop will deal with fundamental aspects of optimization such as complexity in convex and non-convex optimization, relaxation hierarchies, algorithm convergence study. Particular attention will be paid to the resolution of large-scale problems by exploiting various structures including hidden convexity, partial separability or any other form of structure resulting from particular applications. The main subtopics are the following:

  • Composite Optimization : minimization of combinations of smooth functions and convex functions. This area has experienced significant developments stimulated by applications requiring some particular structures: row matrices, sparsity for example.
  • Convex Optimization : this sub-theme covers both fundamental, geometrica and algorothmic aspects around convexity. It has become a fundamental tool, especially for very high dimensional problems related to data processing, around first order methods that nedd to be better understood, accelerated and parallelized for example.
  • Conic Optimization and Hierarchies : the Lasserre Hierarchies have introduced very general relaxation tools with remakarble convergence properties and applications especially for NP-hard problems. This sub-theme will focus on different aspects related to conic programming and the extension the hierarchies of relaxations to other domains (tensor, quantum information, etc.)
  • Nonlinear and Nonconvex Programming : recent efforts have been made to better understand some fundamental aspects around globally convergent algorithms for the optimization of non-convex functions. These advances have given rise to a new generation of algorithms and a perspective of known algorithms. The purpose of this sub-theme is to explore these two aspects: non-convex theory and algorithms.
  • Analysis and exploitation of the geometry. This sub-theme will be an opportunity to present fundamental aspects on the structure of some very large problems.

The « Optimization: Fundamentals and Algorithms dor Structured Problem » workshop will take place at Institut de Recherche en Informatique de Toulouse (IRIT) and is part of the thematic semester « Optimization » organized by Labex CIMI.


Registration to attend the Workshop is now open. Please note that the registration is free but mandatory.

Please note that limited funding is available for student's accommodation. Selected applicants will be notified by Wednesday 2nd of May at the latest (a short CV must be provided).

Register now

Date of the Master Class

28 - 29 June 2018

Confirmed speakers


The workshop will start on Thursday June 28, at 9:00 and will end on Friday June 29 at 16:00. It will take place in the Auditorium at Institut de Recherche en informatique de Toulouse (IRIT). Click here for a Google Map link.

Detailed program (under construction)

Organizing commitee

Sébastien Gadat (UT1), Serge Gratton (IRIT), Aude Rondepierre (IMT), Pierre Weiss (ITAV / IMT).

Scientific commitee

Sébastien Gadat (UT1), Serge Gratton (IRIT), Jean_Bernard Lasserre (LAAS-CNRS), François Malgouyres (IMT), Aude Rondepierre (IMT), Pierre Weiss (ITAV / IMT).



For any questions, please contact us : send an email