Title | ||
---|---|---|
Dealing with categorical and integer-valued variables in Bayesian Optimization with Gaussian processes |
Abstract | ||
---|---|---|
•Bayesian Optimization can be applied to optimization problems with categorical and integer-valued variables.•Existing methods relying on Gaussian processes may get stuck in such a setting.•A transformation of the covariance function is proposed to deal with categorical and integer-valued variables.•Empirical results show that the proposed method outperforms other approaches for Bayesian optimization. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1016/j.neucom.2019.11.004 | Neurocomputing |
Keywords | Field | DocType |
Parameter tuning,Bayesian optimization,Gaussian processes,Integer-valued variables,Categorical variables | Integer,Mathematical optimization,Expected utility hypothesis,Categorical variable,Bayesian optimization,Rounding,Statistical model,Gaussian process,Artificial intelligence,Optimization problem,Mathematics,Machine learning | Journal |
Volume | ISSN | Citations |
380 | 0925-2312 | 6 |
PageRank | References | Authors |
0.46 | 11 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
E. Garrido | 1 | 10 | 2.27 |
Daniel Hernández-Lobato | 2 | 440 | 26.10 |