Learning on Chips

Machine learning is a powerful computer science technique which can derive knowledge from big data, and provides prediction and matching. Since nanometer VLSI CAD problems have extremely high complexity and gigantic data, there has been a surge recently in applying and adapting machine learning techniques in VLSI CAD.


Machine Learning

hotspots 

Selected recent publications:

  • [C57] Hang Zhang, Fengyuan Zhu, Haocheng Li, Evangeline F. Y. Young, Bei Yu, “Bilinear Lithography Hotspot Detection”, ACM International Symposium on Physical Design (ISPD), pp. 7–14, Portland, OR, Mar. 19–22, 2017. (paper) (Best Paper Award)


Deep Learning

hotspots 

Selected recent publications:

  • [C73] Haoyu Yang, Shuhe Li, Yuzhe Ma, Bei Yu, Evangeline F. Y. Young, “GAN-OPC: Mask Optimization with Lithography-guided Generative Adversarial Nets”, ACM/IEEE Design Automation Conference (DAC), San Francisco, CA, June 24–28, 2018. (preprint)



Combinatorial Algorithms in VLSI CAD

Many classical VLSI CAD problems can be extracted and formulated into challenging combinatorial optimization problems. We are heavily working to improve the state-of-the-art.

hotspots 

Selected recent publications are listed as follows:


Detailed Placement

  • [C72] Haocheng Li, Wing-Kai Chow, Gengjie Chen, Evangeline F. Y. Young, Bei Yu, “Routability-Driven and Fence-Aware Legalization for Mixed-Cell-Height Circuits”, ACM/IEEE Design Automation Conference (DAC), San Francisco, CA, June 24–28, 2018. (preprint)


Layout Decomposition


Layer Assignment



Emerging Challenges

Cyber-physical system (CPS) addresses the close interactions and feedback loop between the cyber components such as sensing systems and the physical components such as varying environment and energy systems. Currently we are working on data regression and calibration in smart building environment. On ther other hand, we are also working on improving the security for hardware systems.

emerging topics 

Selected recent publications:

Cyber Physical System


Hardware Security