Conference
Bayesian Optimization for Categorical and Mixed Variables Using a Multinomial Logit Surrogate
| Τίτλος: | Bayesian Optimization for Categorical and Mixed Variables Using a Multinomial Logit Surrogate |
|---|---|
| Συγγραφείς: | Saeed, Muhammad Amir, Candelieri, Antonio |
| Συνεισφορές: | Saeed, M, Candelieri, A |
| Στοιχεία εκδότη: | IT Turin |
| Έτος έκδοσης: | 2026 |
| Συλλογή: | Università degli Studi di Milano-Bicocca: BOA (Bicocca Open Archive) |
| Θεματικοί όροι: | Bayesian optimization, categorical variable, multinomial logit model, black-box optimization, discrete optimization |
| Περιγραφή: | Bayesian optimization is effective for expensive black-box problems, but standard Gaussian-process surrogates are less suitable for categorical and mixed-variable search spaces. We propose MNL-BO, a preference-based Bayesian optimization method that uses a multinomial logit surrogate trained from pairwise comparisons. The model provides interpretable utility estimates for categorical alternatives and supports continuous, discrete, and categorical variables in a unified framework. Experiments on categorical benchmarks, the Traveling Salesman Problem, and mixed-variable pressure vessel design show competitive performance against random search, local search, metaheuristics, and SMAC-inspired tree-based Bayesian optimization baselines. |
| Τύπος εγγράφου: | conference object |
| Περιγραφή αρχείου: | STAMPA |
| Γλώσσα: | English |
| Relation: | ispartofbook:StatTalk 2026: Book of Poster Abstracts; Statalk 2026 - 21 – 22 May 2026; firstpage:1; lastpage:1; numberofpages:1; https://hdl.handle.net/10281/608601 |
| Διαθεσιμότητα: | https://hdl.handle.net/10281/608601 https://sites.google.com/view/statalk-2026/programme?authuser=0 |
| Rights: | info:eu-repo/semantics/openAccess ; license:Altro ; license uri:other-oa |
| Αριθμός Καταχώρησης: | edsbas.28BD1529 |
| Βάση Δεδομένων: | BASE |
| Η περιγραφή δεν είναι διαθέσιμη |