BBOPlace-Bench: Benchmarking Black-Box Optimization for Chip Placement

Ke Xue, Ruo-Tong Chen, Rong-Xi Tan, Xi Lin, Yunqi Shi, Siyuan Xu, Mingxuan Yuan, Chao Qian

Keywords: Black-box optimization, Bayesian optimization, Evolutionary algorithm, Chip placement, EDA

Abstract: Chip placement is a crucial step in modern chip design, because it significantly impacts the subsequent process and the overall quality of the final chip. The application of black-box optimization (BBO) for chip placement has a history of several decades. Nevertheless, early attempts were hampered by immature problem modeling and inefficient algorithm design, resulting in suboptimal placement efficiency and quality compared to the more prevalent analytical methods. Recent advancements in problem modeling and BBO algorithm design have highlighted the effectiveness and efficiency of BBO, demonstrating its potential to achieve state-of-the-art results in chip placement. Despite these advancements, the field lacks a unified benchmark for thoroughly assessing various problem models and BBO algorithms. To address this gap, we propose BBOPlace-Bench, the first benchmark designed for evaluating and developing BBO algorithms specifically for chip placement tasks. BBOPlace-Bench first collects several popular tasks and standardizing their formats, thereby providing uniform and comprehensive information for optimization. Additionally, BBOPlace-Bench includes a wide range of existing BBO algorithms, including simulated annealing, evolutionary algorithms, evolution strategy, and Bayesian optimization, and evaluates their performance across different problem modelings (i.e., permutation, discrete, and mixed search spaces) using various metrics. Furthermore, BBOPlace-Bench offers a flexible framework that allows users to easily implement and test their unique algorithms. BBOPlace-Bench not only provides efficient solutions for chip placement but also expands the practical application scenarios for various BBO algorithms. The code for BBOPlace-Bench is available in the supplementary file.

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: 14043