Henrik A. Friberg: The conic advantage in MINLP скачать в хорошем качестве

Henrik A. Friberg: The conic advantage in MINLP

Best practices in mixed-integer nonlinear programming---each proposed on their own to harvest certain computational advantages---are shown to lead to a mixed-integer general conic optimization framework. The talk continues to discuss a reasonable subset of cones to consider in any software implementation, touching upon convergence properties in Newton methods. Finally, on the open research question of how to break down general nonlinear constraints into their conic atoms, it provides two reformulation techniques capable of handling many of the constraints that users have asked about on public forums. Relevant links: MOSEK modeling cookbook and optimization software: www.mosek.com Small selection of papers to read if you are interested in mixed-integer conic optimization: Günlük and Linderoth (2012): Perspective reformulation and applications, Mixed Integer Nonlinear Programming, 154. Lodi, Tanneau and Vielma (2020): Disjunctive cuts for Mixed-Integer Conic Optimization. Belotti, Goez, Polik, Ralphs and Terlaky (2017): A complete characterization of disjunctive conic cuts for mixed integer second order cone optimization, Discrete Optimization, 24. Shahabsafa, Goez and Terlaky (2018): On pathological disjunctions and redundant disjunctive conic cuts. Operations Research Letters, 46(5). Coey, Lubin, Vielma (2020): Outer approximation with conic certificates for mixed-integer convex problems, Mathematical Programming Computation, 12. ~ This talk was submitted to MINLP Virtual Workshop 2021 (https://optimisation.doc.ic.ac.uk/min...) Connect with the Computational Optimisation Group at Imperial College London online... Subscribe to the CogImperial YouTube channel for more research related content:    / @cogimperial   Follow us on Twitter for news about our research group:   / cogimperial   Visit our official site for more information: https://optimisation.doc.ic.ac.uk/