EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Machine Learning-Driven Chip Design: Design Methodology and EDA Evolution in the Post-Moore Era

Jiahao Tian

MATEC Web of Conferences · 2026 · 424 · 04014

DOI: 10.1051/matecconf/202642404014 ↗

Kurz gefasst

Fachartikel in der Zeitschrift MATEC Web of Conferences, online erschienen am 19.08.2026. Verfasst von Jiahao Tian. Die Arbeit nennt 10 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The stop of Moore's Law is making chip design face the difficulty to get breaks past the super-Pareto boundary in power, performance, and area (PPA). This thesis systematically carries out exploration on design methods that are based on artificial intelligence and machine learning, with the focus being put on the intelligent evolution of electronic design automation (EDA) tools. By means of reinforcement learning and methods which depend on data, artificial intelligence has obtained remarkable outcomes in domains such as multi-dimensional PPA trade-off searching, multiple-goal optimization in architecture design, physical design automatic processing, timing and power optimization, verification speed increase, and large language model-based (LLM) register transfer level (RTL) generation, therefore it greatly promotes iteration efficiency and PPA performance. Although the difficulties like data dependence and multiple-target optimization are still existing, this paper has discussed the potential application situations of AI-EDA cooperation design on system-level optimization and self-adaptive tool chains. This research provides a crucial theoretical and practical foundation for breaking through the PPA bottleneck in semiconductor design beyond Moore's Law.

Zitieren

APA

Tian, J. (2026). Machine Learning-Driven Chip Design: Design Methodology and EDA Evolution in the Post-Moore Era. MATEC Web of Conferences, 424, 04014. https://doi.org/10.1051/matecconf/202642404014

Vancouver

Tian J. Machine Learning-Driven Chip Design: Design Methodology and EDA Evolution in the Post-Moore Era. MATEC Web of Conferences. 2026;424:04014. doi:10.1051/matecconf/202642404014

BibTeX
@article{Tian2026,
  author = {Tian, Jiahao},
  title = {Machine Learning-Driven Chip Design: Design Methodology and EDA Evolution in the Post-Moore Era},
  journal = {MATEC Web of Conferences},
  year = {2026},
  volume = {424},
  pages = {04014},
  doi = {10.1051/matecconf/202642404014}
}
Kostenlos starten →