DesignCon 2022 Schedule Builder

Minsu Kim received a dual degree of B.S. in Mathematics and Computer Science from the Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea, in 2020. He will recieve (Feb 2022) M.S. degree in Electrical Engineering from the KAIST. He is now pursuing Ph.D degree, focused on operational research (OR) and reinforcement learning (RL). His current research interests include Reinforcement Learning (RL) application to channel routing considering signal integrity (SI). He achieved the best student paper in IEEE EDAPS 2020. He is now nominated for the best paper award in DesignCon 2021 for his paper "Neural Language Model Enables Extremely Fast & Robust Routing on Interposer". He published 15 scientific papers including both top-tier machine learning paper (NeurIPS) and SI/PI papers (EPEPS; EDAPS; DesignCon).

Presenting:

• A Processing-In-memory on High-bandwidth Memory (PIM-HBM): Impact of Interconnect Channels on System Performance in 2.5D/3D IC
• Data-Efficient Machine Learning Method for Hardware Routing: Signal Integrity Demonstration to 2PDC and PCIe 6.0
• Deep Reinforcement Learning-based Channel-flexible Equalization Scheme: An Application to High Bandwidth Memory
• Imitate Expert Policy & Learn Beyond: A Practical PDN Optimizer by Imitation Learning
• Imitation Learning with Bayesian Exploration (IL-BE) for Signal Integrity (SI) of PAM-4 based High-speed Serial Link: PCIe 6.0
• Learning Super-scale Microbump Pin Assignment Optimization for Real-world PCB Design with Graph Representation
• Thermal Transmission Line: Smoothing Thermal Gradients & Lowering Temperature for Signal Integrity Improvement of HBM & 2.5D ICs