Event
Ph.D. Research Proposal Exam: Jiaqi Yin
Friday, May 9, 2025
10:00 a.m.
AVW 2328
Maria Hoo
301 405 3681
mch@umd.edu
ANNOUNCEMENT: Ph.D. Research Proposal Exam
Name: Jiaqi Yin
Committee:
Professor Cunxi Yu
Professor Ang Li
Professor Shuvra S. Bhattacharyya
Date/time: Friday, May 9th from 10:00 AM to 11:59 AM
Location: AVW 2328 (ECE)
Title: Chasing Optimal at Speed: Intelligent Combinatorial Problem Solving in Computing Systems Design
Abstract: Combinatorial optimization problems are widespread in electronic design automation (EDA), scheduling, and system design, yet their computational complexity often makes traditional methods impractical for large-scale scenarios. My dissertation focuses on optimizing key challenges in this computing system design, including scheduling for high-level synthesis (HLS), symbolic reasoning for EDA, large integer multiplication design for FPGA design, and formal verification. To tackle these effectively, I propose a multifaceted framework that leverages domain-specific heuristics tailored to each problem structure, reinforcement learning for mimicking optimal solutions, exact methods for guaranteed solutions, equality saturation for exploring optimization spaces, and comprehensive combinations of these techniques. Through rigorous theoretical analysis and extensive practical implementations, I demonstrate that this approach significantly improves speed and scalability over conventional solvers.
