[Opt-Net] GIAN Course on Advances in Mixed Integer Nonlinear Optimization -- IIT Bombay

Pietro Luigi Belotti pietro.belotti at polimi.it
Tue Jan 2 16:20:18 CET 2024


This course is part of the Global Initiative of Academic Networks (GIAN) program.

Venue: Victor Menezes Convention Centre, IIT Bombay

Dates: 15-25th January, 2024

Many design, planning and decision problems arising in engineering, sciences, finance, and statistics can be mathematically modeled as Mixed-Integer Nonlinear Optimization (MINLO) problems. The last two decades have seen a phenomenal growth in the development of theory, algorithms and computational tools for MINLO. Riding on this growth, MINLO has found many applications, and is being used to solve practical problems in various domains.

The 10-day (about 50 hours) course will start with a gentle introduction to MINLO models and motivating practical applications. The next part of the course covers the branch-and-cut paradigm which is central to solving these problems followed by methods for convex MINLOs and nonconvex MINLOs. Heuristic search, cutting plane techniques, reformulation and presolving methods will be covered. The course concludes with discussions on the future trends of MINLO. Tutorials for practice problems and exercises will be conducted each day. Hands-on computational exercises on modeling and using state-of-the-art software will help participants understand practical ways of solving MINLOs.

Industry practitioners have the option of registering for a shorter module spanning six days (15-20 January). It will focus on introduction to MINLO, modeling, software and methods for convex MINLO.

Faculty:
    Sven Leyffer, Argonne National Lab, Chicago
    Pietro Belotti, Politecnico di Milano
    Ashutosh Mahajan, IEOR, IIT Bombay

More information at https://www.ieor.iitb.ac.in/minlo23

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.zib.de/pipermail/opt-net/attachments/20240102/6fea2d66/attachment.html>


More information about the Opt-Net mailing list