Algorithmic exploration becomes easier in electronic design automation research

When Algorithmic Exploration Becomes Cheap: A Case Study of Agentic Research in EDA

Hardware ArchitectureArtificial Intelligence

Summary

The authors studied how algorithms that explore new ideas can work with less human help in electronic design automation (EDA), the field concerned with designing computer chips and circuits. They found that these algorithms, even when run by researchers less experienced in the area, can create new mathematical tools and improve existing ones. The team also analyzed thousands of research papers and noticed that most papers focus on well-defined computational problems. This means EDA provides a rich environment for automated algorithm development. The authors highlight new challenges for verifying and valuing research outputs when producing results is easier than carefully checking them.

What this means in practice

  • For eda tool developers: Use automated agents to test and refine new algorithmic ideas in EDA tools that were previously too time-consuming to explore.
  • For software verification teams: Adopt new procedures to verify algorithmic research outputs when automation makes producing results faster than manual checking.

Authors

Keren Zhu, Yu Deng, Xiaoyu Hao, Liwen Jiang, Zijian Jiang, Cunqing Lan, Boxiang Song, Pujun Su, Yaojia Wang

Abstract

As EDA researchers, we conducted eight deliberate trials of agentic algorithm exploration, selecting several topics outside our areas of depth. One faculty member and seven students participated, including students without publication experience. With limited intervention in the algorithms, agents developed mathematical constructions, analyzed existing tools, and implemented improvements; some efforts fell short of their practical goals. We also used AI to collect, classify, and analyze 8,420 papers from four EDA conferences and two journals over 2022-2026. Among 2,380 primary-core papers, we classified 97.7% from titles and abstracts as computationally closed, including work on new formulations. Together, these observations suggest that much of EDA offers an executable environment for increasingly accessible algorithm research. We see an opportunity for tool developers to investigate ideas they previously lacked time to pursue. We also ask how EDA should validate and reward research when results become easier to produce than to examine, and what papers and venue labels will continue to tell us about a contribution.
View PDFOpen arXiv