Eric Qin
Papers
1
Total Citations
462
H-Index
1
About
Eric Qin is a leading researcher in computer architecture, with a primary focus on accelerating deep neural network (DNN) training through novel hardware designs. His most influential contribution is the SIGMA architecture, a sparse and irregular GEMM accelerator featuring flexible interconnects—a seminal work that has garnered 462 citations since 2020. SIGMA directly addresses the critical challenge of efficiently handling sparsity in DNN training, moving beyond traditional dense matrix multiplication accelerators to enable significant performance and energy gains. This work has shaped how the field approaches hardware-software co-design for modern, irregular neural network workloads. Beyond SIGMA, Qin’s research spans the broader landscape of efficient computing for vision, speech, and robotics, consistently targeting the intersection of algorithm innovation and circuit-level implementation. His contributions are widely recognized as foundational for next-generation training accelerators, making him a key voice in the push toward more flexible and performant deep learning hardware.
Research Focus
Key Achievements
Top Papers
- 1