Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2504.17801 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912346280034304 |
|---|---|
| author | Yao, Xufeng Jiang, Jiaxi Zhao, Yuxuan Liao, Peiyu Lin, Yibo Yu, Bei |
| author_facet | Yao, Xufeng Jiang, Jiaxi Zhao, Yuxuan Liao, Peiyu Lin, Yibo Yu, Bei |
| contents | Optimization algorithms are widely employed to tackle complex problems, but designing them manually is often labor-intensive and requires significant expertise. Global placement is a fundamental step in electronic design automation (EDA). While analytical approaches represent the state-of-the-art (SOTA) in global placement, their core optimization algorithms remain heavily dependent on heuristics and customized components, such as initialization strategies, preconditioning methods, and line search techniques. This paper presents an automated framework that leverages large language models (LLM) to evolve optimization algorithms for global placement. We first generate diverse candidate algorithms using LLM through carefully crafted prompts. Then we introduce an LLM-based genetic flow to evolve selected candidate algorithms. The discovered optimization algorithms exhibit substantial performance improvements across many benchmarks. Specifically, Our design-case-specific discovered algorithms achieve average HPWL improvements of \textbf{5.05\%}, \text{5.29\%} and \textbf{8.30\%} on MMS, ISPD2005 and ISPD2019 benchmarks, and up to \textbf{17\%} improvements on individual cases. Additionally, the discovered algorithms demonstrate good generalization ability and are complementary to existing parameter-tuning methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_17801 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Evolution of Optimization Algorithms for Global Placement via Large Language Models Yao, Xufeng Jiang, Jiaxi Zhao, Yuxuan Liao, Peiyu Lin, Yibo Yu, Bei Neural and Evolutionary Computing Artificial Intelligence Optimization algorithms are widely employed to tackle complex problems, but designing them manually is often labor-intensive and requires significant expertise. Global placement is a fundamental step in electronic design automation (EDA). While analytical approaches represent the state-of-the-art (SOTA) in global placement, their core optimization algorithms remain heavily dependent on heuristics and customized components, such as initialization strategies, preconditioning methods, and line search techniques. This paper presents an automated framework that leverages large language models (LLM) to evolve optimization algorithms for global placement. We first generate diverse candidate algorithms using LLM through carefully crafted prompts. Then we introduce an LLM-based genetic flow to evolve selected candidate algorithms. The discovered optimization algorithms exhibit substantial performance improvements across many benchmarks. Specifically, Our design-case-specific discovered algorithms achieve average HPWL improvements of \textbf{5.05\%}, \text{5.29\%} and \textbf{8.30\%} on MMS, ISPD2005 and ISPD2019 benchmarks, and up to \textbf{17\%} improvements on individual cases. Additionally, the discovered algorithms demonstrate good generalization ability and are complementary to existing parameter-tuning methods. |
| title | Evolution of Optimization Algorithms for Global Placement via Large Language Models |
| topic | Neural and Evolutionary Computing Artificial Intelligence |
| url | https://arxiv.org/abs/2504.17801 |