Abstract
Placement is a crucial and time-consuming step in modern very-large scale integration (VLSI) physical design. In this survey, we provide a comprehensive review of existing machine learning techniques for placement studies and categorize them as follows: intelligent placement and kernelized placement. Intelligent placement that combines machine-learning algorithms and traditional placement methods will continue to be a trending topic in academic research. Kernelized placement frameworks that can significantly accelerate solving multi-objective placement problems using AI hardware have also received considerable attention recently. We discuss the advantages and drawbacks of intelligent placement and kernelized placement and highlight future trends in machine learning techniques for placement.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have