Library

Solving Black-Box Optimization Challenge via Learning Search Space Partition for Local Bayesian Optimization

Black-box optimization is one of the vital tasks in machine learning, since\nit approximates real-world conditions, in that we do not always know all the\nproperties of a given system, up to knowing almost nothing but the results.\nThis paper describes our approach to solving the black-box optimization\nchallenge at NeurIPS 2020 through learning search space partition for local\nBayesian optimization. We describe the task of the challenge as well as our\nalgorithm for low budget optimization that we named \\texttt{SPBOpt}. We\noptimize the hyper-parameters of our algorithm for the competition finals using\nmulti-task Bayesian optimization on results from the first two evaluation\nsettings. Our approach has ranked third in the competition finals.\n

cs.LGMachine Learning and Data ClassificationAdvanced Bandit Algorithms ResearchAdvanced Neural Network ApplicationsComputer Science
Source

Paper

Similar papers

Bayesian Optimization is Superior to Random Search for Machine Learning Hyperparameter Tuning: Analysis of the Black-Box Optimization Challenge 2020Ryan Turner, David Eriksson, Michael McCourt et al.2020 In Library
Bayesian Optimization is Superior to Random Search for Machine Learning Hyperparameter Tuning: Analysis of the Black-Box Optimization Challenge 2020Ryan Turner, David Eriksson, Michael McCourt et al.2021 · arXiv (Cornell University) · 404 citations In Library
Adaptive Local Bayesian Optimization Over Multiple Discrete VariablesTaehyeon Kim, Jaeyeon Ahn, Nakyil Kim et al.2020 · arXiv (Cornell University) In Library
Learning Search Space Partition for Black-box Optimization using Monte Carlo Tree SearchLinnan Wang, Rodrigo Fonseca, Yuandong Tian2020 · arXiv (Cornell University) · 159 citations In Library
Machine Learning Algorithms for Improving Black Box Optimization SolversM. Kimiaei, V. Kungurtsev2025 · arXiv.org · 2 citations In Library
Scalable neural network-based blackbox optimizationPavankumar Koratikere, Leifur Leifsson2025 · Structural And Multidisciplinary Optimization · 0 citations In Library
Think Global and Act Local: Bayesian Optimisation over High-Dimensional Categorical and Mixed Search SpacesXingchen Wan, Loc Nguyen, Huong Ha et al.2021 · arXiv (Cornell University) · 86 citations In Library
Learning search spaces for Bayesian optimization: Another view of hyperparameter transfer learningValerio Perrone, Huibin Shen, M. Seeger et al.2019 · Neural Information Processing Systems · 118 citations In Library
B2Opt: Learning to Optimize Black-box Optimization with Little BudgetXiaobin Li, Kai Wu, Xiaoyu Zhang et al.2023 · arXiv (Cornell University) · 24 citations In Library
Predicting the utility of search spaces for black-box optimization: a simple, budget-aware approachSetareh Ariafar, Justin Gilmer, Zachary Nado et al.2021 · arXiv (Cornell University) · 2 citations In Library
© 2026 NYSGPT2525 LLC