Flexible Multiple-Objective Reinforcement Learning for Chip Placement
Abstract
Recently, successful applications of reinforcement learning to chip placement have emerged. Pretrained models are necessary to improve efficiency and effectiveness. Currently, the weights of objective metrics (e.g., wirelength, congestion, and timing) are fixed during pretraining. However, fixed-weighed models cannot generate the diversity of placements required for engineers to accommodate changing requirements as they arise. This paper proposes flexible multiple-objective reinforcement learning (MORL) to support objective functions with inference-time variable weights using just a single pretrained model. Our macro placement results show that MORL can generate the Pareto frontier of multiple objectives effectively.
Cite
@article{arxiv.2204.06407,
title = {Flexible Multiple-Objective Reinforcement Learning for Chip Placement},
author = {Fu-Chieh Chang and Yu-Wei Tseng and Ya-Wen Yu and Ssu-Rui Lee and Alexandru Cioba and I-Lun Tseng and Da-shan Shiu and Jhih-Wei Hsu and Cheng-Yuan Wang and Chien-Yi Yang and Ren-Chu Wang and Yao-Wen Chang and Tai-Chen Chen and Tung-Chieh Chen},
journal= {arXiv preprint arXiv:2204.06407},
year = {2022}
}
Comments
A short version of this article is published in DAC'22:LBR (see ACM DOI 10.1145/3489517.3530617)