This repository contains the source code of OPT-GAN and examples of how to use it to optimize given black-box problems, such as 'Conformal Bent Cigar' or 'NEEP'.
This folder includes the code for OPT-GAN and an example of how to use it to optimize the 'Conformal Bent Cigar' problem. If you wish to use OPT-GAN for optimizing other black-box problems, you will need to replace the corresponding content with the target problem.
if func == "Conformal_Bent_Cigar":
problem = Problem(problem_Conformal_Bent_Cigar) #define the target black-box problem
upper = 5 # the up bound of the black-box problem's solution
lower = -5 # the low bound of the black-box problem's solution
To specify the parameters of OPT-GAN while optimizing a black-box problem, you can modify them when running the Python file as follows:
python experiment_Conformal_Bent_Cigar.py --gamma=1.5 --lambda2=0.3 --pop=30 --optset=150 --func_dim=2 --func_ins=0 --func_id=Conformal_Bent_Cigar --func_alg=OPT-GAN --maxfes=50000
The folder also contains an example of how to use OPT-GAN to optimize the 'NEEP' problem.
If our work is helpful to you, please kindly cite our paper as:
@article{Lu_Ning_Liu_Sun_Zhang_Yang_Wang_2023,
title = {OPT-GAN: A Broad-Spectrum Global Optimizer for Black-Box Problems by Learning Distribution},
volume = {37},
url = {https://ojs.aaai.org/index.php/AAAI/article/view/26468},
doi = {10.1609/aaai.v37i10.26468},
number = {10},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
author = {Lu, Minfang and Ning, Shuai and Liu, Shuangrong and Sun, Fengyang and Zhang, Bo and Yang, Bo and Wang, Lin},
year = {2023},
month = {Jun.},
pages = {12462-12472}
}