. Hansen, P. & Jaumard, B. Horst, R. & Pardalos, P. M., editors. Lipschitz Optimization, pages 407–493. Springer US, Boston, MA, 1995.
Lipschitz Optimization [link]Paper  doi  abstract   bibtex   
Lipschitz optimization solves global optimization problems in which the objective function and constraint left-hand sides may be given by oracles (or explicitly) and have a bounded slope. The problems of finding an optimal solution, an $ε$-optimal one, all optimal solutions, and a small volume enclosure of all optimal solutions within hyperrectangles (possibly containing only a-optimal points) are investigated. In the univariate case, necessary conditions for finite convergence are studied as well as bounds on the number of iterations and convergence of e-optimal algorithms. Methods of Piyayskii, Evtushenko, Timonov, Schoen, Galperin, Shen and Zhu, and Hansen, Jaumard and Lu are discussed and compared experimentally. The same is done in the multivariate case for the algorithms of Piyayskii; Mladineo; Jaumard, Herrmann and Ribault; Pinter; Galperin; Meewella and Mayne; Wood; Gourdin, Hansen and Jaumard.
@InBook{Hansen1995,
    author      = {Hansen, Pierre and Jaumard, Brigitte},
    title       = {Lipschitz Optimization},
    address     = {Boston, MA},
    booktitle   = {Handbook of Global Optimization},
    doi         = {10.1007/978-1-4615-2025-2_9},
    editor      = {Horst, Reiner and Pardalos, Panos M.},
    isbn        = {978-1-4615-2025-2},
    pages       = {407--493},
    publisher   = {Springer US},
    url         = {https://doi.org/10.1007/978-1-4615-2025-2_9},
    year        = {1995},
    abstract    = {Lipschitz optimization solves global optimization problems in which the objective function and constraint left-hand sides may be given by oracles (or explicitly) and have a bounded slope. The problems of finding an optimal solution, an $\epsilon$-optimal one, all optimal solutions, and a small volume enclosure of all optimal solutions within hyperrectangles (possibly containing only a-optimal points) are investigated. In the univariate case, necessary conditions for finite convergence are studied as well as bounds on the number of iterations and convergence of e-optimal algorithms. Methods of Piyayskii, Evtushenko, Timonov, Schoen, Galperin, Shen and Zhu, and Hansen, Jaumard and Lu are discussed and compared experimentally. The same is done in the multivariate case for the algorithms of Piyayskii; Mladineo; Jaumard, Herrmann and Ribault; Pinter; Galperin; Meewella and Mayne; Wood; Gourdin, Hansen and Jaumard.}
}

Downloads: 0