Furen Zhuang, Partha Pratim Kundu, Sezin Kircali Ata, Y H, Mihai Dragos Rotaru, Rahul Dutta, & Ashish James (2025). Bayesian Optimization for Chiplet Placement on 2.5D System in Package Platforms: Tackling Rotation and Constraint Challenges. In AI4X 2025 International Conference.
Abstract:
In System-in-Package (SiP) design, current Bayesian optimization (BO)-based floor-planning methods fail to address chiplet rotation and infeasibly large placements, leading to inefficiencies in performance and optimization. Our work introduces a novel rotation- and constraint-aware BO approach for multi-objective chiplet placement, and effectively overcomes these limitations for enhanced performance.
License type:
Publisher Copyright
Funding Info:
This research / project is supported by the A*STAR - Manufacturing, Trade, and Connectivity Programmatic Fund
Grant Reference no. : M23M3b0064