Overview

Implementation code for paper MOCA-HESP: Meta High-dimensional Bayesian Optimization for Combinatorial and Mixed Spaces via Hyper-ellipsoid Partitioning published in the 28th European Conference on Artificial Intelligence (ECAI 2025). The manuscript will be made available soon.

If you find our paper or this repo to be useful for your research, please consider citing:

@inproceedings{ngo2025mocahesp,
  title={MOCA-HESP: Meta High-dimensional Bayesian Optimization for Combinatorial and Mixed Spaces via Hyper-ellipsoid Partitioning},
  author={Ngo, Lam and Ha, Huong and Chan, Jeffrey and Zhang, Hongyu},
  booktitle={European Conference on Artificial Intelligence},
  year={2025}
}

Requirement and dependencies

• Install Anaconda/Miniconda.
• Install env from mocahesp.yml file by running the following line via conda:

conda env create -f mocahesp.yml

• (Optional) If you want to run antibody benchmark, run the following line:

bash install-antibody-lib.sh

• (Optional) If you want to run Cellular Network Optimization becnhmark: download all files in power_maps folder at https://github.com/Ryandry1st/CCO-in-ORAN/tree/main/cco_standalone_icassp_2021/data/power_maps to _benchmark_problems/cco/powermaps

Getting started

The main file is test-mocahesp.py. To run MOCA-HESP-BO for the non-shifted Ackley20c problem, use the following syntax:

python test-mocahesp.py --solver bo -f ackley20c -n 400 --seed 1

Please see the following table for all arguments, as well as their possible values.

Flag	Description
-f	A string specifying objective function's name. Possible options include categorical problems (ackley20c, maxsat28, maxsat125, labs, antibody) and mixed problems (ackley53m, cco, svm). To add more benchmark problems, follow the templates in mocahesp/function_wrapper.py.
-n	An integer specifying the iteration budget.
--solver	(Optional) A string specifying the BO solver to be integrated. Possible values include bo, casmo and bounce. Default is bo
--shifted	(Optional) A flag specifying whether to use the shifted version of benchmark problem. If not set, the original benchmark problems are used. Default is original problem.
--n_init	(Optional) An integer specifying the number of initialization. Recommend to be higher than the maximum number of choices across all categorical/ordinal variables. Default is 20.
--seed	(Optional) An integer to specify the random seed for random number generator. Default is 1.

Results from paper

To produce the main experimental results in the paper (Fig. 2), run the bash file experiments-from-paper.sh.

Acknowledgements

This code implementation uses materials from the following public repositories to implemenent the subspace embedding. We thank the respective repository maintainers.

1. CMA-ES: Hansen, N., & Ostermeier, A. (2001). Completely derandomized self-adaptation in evolution strategies. Evolutionary computation, 9(2), 159-195. Code repo: https://github.com/CMA-ES/pycma
2. CASMOPOLITAN: Wan, Xingchen, et al. "Think Global and Act Local: Bayesian Optimisation over High-Dimensional Categorical and Mixed Search Spaces." International Conference on Machine Learning. PMLR, 2021.. Code repo: https://github.com/xingchenwan/Casmopolitan
3. Bounce: Papenmeier, Leonard, Luigi Nardi, and Matthias Poloczek. "Bounce: Reliable high-dimensional Bayesian optimization for combinatorial and mixed spaces." Advances in Neural Information Processing Systems 36 (2023): 1764-1793. Code repo: https://github.com/LeoIV/bounce