GLIS/GLISp/C-GLISP - Global Optimization, Radial Basis Functions, Inverse Distance Weighting, Preference Learning

<!DOCTYPE HTML>  <html> <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> <link href="http://fonts.googleapis.com/css?family=Oswald|Lobster|Average" rel="stylesheet" type="text/css" /> <!see https://developers.google.com/fonts/docs/getting_started>  <script async src="https://www.googletagmanager.com/gtag/js?id=UA-16599435-7"></script> <script> window.dataLayer = window.dataLayer || []; function gtag(){dataLayer.push(arguments);} gtag('js', new Date()); gtag('config', 'UA-16599435-7'); </script> <title>GLIS/GLISp/C-GLISP - Global Optimization, Radial Basis Functions, Inverse Distance Weighting, Preference Learning</title> <meta charset="utf-8" /> <meta name="viewport" content="width=device-width, initial-scale=1, user-scalable=no" /> <link rel="stylesheet" href="../assets/css/main.css" /> </head> <body class="is-preload">  <div id="content"> <div class="inner">  <article class="box post post-excerpt"> <header>  <div class="container"> <div class="text"> <large>GLIS / GLISp / C-GLISP</large><br> </div> <div class="image"> <img src="images/glis-logo.png" alt="GLIS"> </div> </div> </header> <p> <strong>Package description</strong> </p> <p>GLIS is a package for finding the global (GL) minimum of a function that is expensive to evaluate, possibly under constraints, using inverse (I) distance weighting and surrogate (S) radial basis functions. </p> <p>The package implements two main algorithms: </p> <p> <b><font color="#990000">GLIS</font></b>: Global optimization based on inverse distance weighting and radial basis function surrogates, a method for global optimization of a function <i>f</i> that is possibly expensive to evaluate. The algorithm is based on the following paper: </p><p> <font color="#004C99">[1] A. Bemporad, "<a href="http://cse.lab.imtlucca.it/~bemporad/publications/papers/coap-glis.pdf">Global optimization via inverse weighting and radial basis functions</a>," Computational Optimization and Applications, vol. 77, pp. 571–595. </font> </p> <b><font color="#990000">GLISp</font></b>: Global optimization of a function <i>f</i> whose value cannot be evaluated but, given two points <i>x</i>,<i>y</i>, it is possible to query whether <i>f(x)</i> is better than <i>f(y)</i> (<i>preference-based optimization</i>). The algorithm is based on the following paper: </p><p> <font color="#004C99">[2] A. Bemporad, D. Piga, "<a href="http://cse.lab.imtlucca.it/~bemporad/publications/papers/mlj_glisp.pdf">Active preference learning based on radial basis functions</a>," Machine Learning, vol. 110, no. 2, pp. 417-448, 2021. Available on <a href="https://arxiv.org/pdf/1909.13049">arXiv:1909.13049</a>.</font> </p> The package also includes <b><font color="#990000">C-GLISp</font></b>, an extension of GLIS and GLISp to handle unknown constraints. The algorithm is based on the following paper: </p><p> <font color="#004C99">[3] M. Zhu, D. Piga, A. Bemporad, "<a href="http://cse.lab.imtlucca.it/~bemporad/publications/papers/ieeecst-c-glisp.pdf">C-GLISp: Preference-based global optimization under unknown constraints with applications to controller calibration</a>,” 2021, IEEE Trans. Contr. Systems Technology, 2021, In press. Also available on arXiv at <a href="https://arxiv.org/pdf/2106.05639">arxiv:2106.05639</a>.</font> </p> <p> This software is distributed without any warranty. Please cite the above papers if you use this software. </p> <p><strong>Download</strong></p> <p> <i>Python version: </i>[<a href="https://github.com/bemporad/GLIS">Source</a>]</p> <p> <strong style="align: center; color:rgb(51, 102, 153);background-color:rgb(250, 250, 250);border-width:1px; border-style:solid; border-color:rgb(200,200,200); padding: 1em; font-family:monospace">pip install glis</strong></p> </p> <p><i>MATLAB version:</i> <a href="https://github.com/bemporad/GLIS_MATLAB">GitHub repo</a></p> <p><i>Previous MATLAB/Python versions:</i></p> <p> <a href="download/glis_v3.0.zip">GLIS v3.0 - June 9, 2021</a><br> (unknown constraints introduced in GLIS and GLISP by Mengjia Zhu)</p> <p> <a href="download/glis_v2.4.zip">GLIS v2.4 - January 12, 2021</a><br> (bugs fixed in PYTHON version of GLISp by Mengjia Zhu)</p> <p> <a href="download/glis_v2.3.zip">GLIS v2.3 - December 20, 2020</a><br> (minor changes in MATLAB version)</p> <p> <a href="download/glis_v2.2.zip">GLIS v2.2 - January 22, 2020</a><br> (added support for Genetic Algorithm method from MATLAB Global Optimization Toolbox. Minor change in GLISp demo - MATLAB version only)</p> <p> <a href="download/glis_v2.1.zip">GLIS v2.1 - October 18, 2019</a><br>(file and function names changed to GLIS/GLISp)</p> <p> <a href="download/glis_v2.0.1.zip">GLIS v2.0.1 - September 29, 2019</a></p> <p> <a href="download/glis_v2.0.zip">GLIS v2.0 - September 28, 2019</a><br> (added preference-based optimization functions. Preference-based Bayesian optimization by D. Piga) </p> <p> <a href="download/glis_v1.1.1.zip">GLIS v1.1.1 - September 2, 2019</a><br> </p> <p> <a href="download/glis_v1.1.zip">GLIS v1.1 - August 3, 2019</a><br> (Python version improved by M. Forgione) </p> <p> <a href="download/glis_v1.0.2.zip">GLIS v1.0.2 - July 6, 2019</a> </p> <p> <a href="download/glis_v1.0.1.zip">GLIS v1.0.1 - July 4, 2019</a> </p> <p> <a href="download/glis_v1.0.zip">GLIS v1.0 - June 15, 2019</a> </p> <p><br><br>Last update: March 3, 2023</p> </article> </div> </div>  <div id="sidebar">  <h1 id="logo"><a href="#"></a></h1>  <nav id="nav"> <ul> <li><a href="../index.html">Home</a></li> <li><a href="http://cse.lab.imtlucca.it/~bemporad/publications/">Publications</a></li> <li><a href="../software.html">Software</a></li> <li><a href="../teaching.html">Teaching</a></li> <li><a href="../talks.html">Talks</a></li> <li><a href="../projects.html">Projects</a></li> <li><a href="../events.html">Events</a></li> <li><a href="../biography.html">About me</a></li> </ul> </nav>  <ul id="copyright"> <li>© A. Bemporad, 2023.</li><li>Web template: <a href="http://html5up.net">HTML5 UP</a></li> </ul> </div>  <script src="../assets/js/jquery.min.js"></script> <script src="../assets/js/browser.min.js"></script> <script src="../assets/js/breakpoints.min.js"></script> <script src="../assets/js/util.js"></script> <script src="../assets/js/main.js"></script> </body> </html>

CINXE.COM

GLIS/GLISp/C-GLISP - Global Optimization, Radial Basis Functions, Inverse Distance Weighting, Preference Learning