Relation Augmented Preferential Bayesian Optimization via Preference Propagation
Keywords: Dueling Optimization, Preference Propagation, Data Augmentation
Abstract: In black-box optimization, when directly evaluating the function values of solutions is very costly or infeasible, access to the objective function is often limited to comparing pairs of solutions, which yields dueling black-box optimization. Dueling optimization is solely based on pairwise preferences, and thus notably reduces cost compared with function value based methods such as Bayesian optimization. However, an optimization performance gap obviously exists between dueling based and function value based methods. This is mainly due to that most existing dueling optimization methods do not make full use of the pairwise preferences collected. To fill this gap, this paper proposes relation augmented preferential Bayesian optimization (RAPBO) via preference propagation. By considering solution similarity, RAPBO aims to uncover the potential preferential relations between solutions within different preferences through the proposed preferential relation propagation technique. Specifically, RAPBO first clusters solutions using a Gaussian mixture model. After obtaining the solution set with the highest intra-cluster similarity, RAPBO utilizes a directed hypergraph to model the potential relations between solutions, thereby realizing relation augmentation. Extensive experiments are conducted on both synthetic functions and real-world tasks such as motion control and spacecraft trajectory optimization. The experimental results disclose the satisfactory accuracy of augmented preferences in RAPBO, and show the superiority of RAPBO compared with existing dueling optimization methods. Notably, it is verified that, under the same evaluation cost budget, RAPBO is competitive with or even surpass the function value based Bayesian optimization methods with respect to optimization performance. The codes can be found in https://anonymous.4open.science/r/RAPBO-E15F.
Primary Area: optimization
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 13770
Loading