GNN-Opt: Enhancing Automated Circuit Design Optimization with Graph Neural Networks

Kazuya YAMAMOTO; Nobukazu TAKAI

doi:10.1587/transfun.2024GCL0001

Special Section on Analog Circuit Techniques and Related Topics

GNN-Opt: Enhancing Automated Circuit Design Optimization with Graph Neural Networks

Kazuya YAMAMOTO, Nobukazu TAKAI

著者情報

キーワード: analog integrated circuits design, design automation, machine learning, deep reinforcement learning

ジャーナルフリー

2025 年 E108.A 巻 5 号 p. 687-689

DOI https://doi.org/10.1587/transfun.2024GCL0001

Browse “Advance online publication” version

詳細

抄録

We introduce “GNN-Opt,” a method that adapts “DNN-Opt’s” machine learning approach for analog circuit design, particularly op-amp sizing. By utilizing Graph Neural Networks, GNN-Opt demonstrates superior efficiency in transferring design knowledge across similar topologies, accelerating to achieve higher FoM with same simulation times.

責任著者(Corresponding author)

J-STAGEへの登録はこちら（無料）