17:00 PM

ECE PhD Student Wins Prestigious Award for Recent Paper

PhD Student Yue Tang received the RAY Award for her paper that assists in on-device learning

Yue Tang, PhD student at the University of Pittsburgh Swanson School of Engineering, is already making a name for herself. 

For her paper, “EF-Train: Enable Efficient On-Device CNN Training on FPGA through Data Reshaping for Online Adaptation or Personalization,” Tang received the ACM TODAES Rookie Author of the Year (RAY) from the ACM Transactions on Design Automation of Electronic Systems (TODAES) journal. 

“I am just really thrilled to have received this award,” Tang said. 

The award was announced at this year’s Design Automation Conference (DAC), the premier conference on electronic design automation, where she was one of six women to win. 

Yue Tang and team

Tang said she was thankful for her “wonderful” co-authors that helped bring the paper to life. 

  • Dr. Xinyi Zhang, Research Engineer at Meta, also an alumni of Pitt
  • Peipei Zhou, Assistant Professor of Electrical and Computer Engineering at Pitt 
  • Jingtong Hu, Associate Professor of Electrical and Computer Engineering, William Kepler Whiteford Fellow, and Tang’s advisor at Pitt 

Hu said Tang was both an “innovative and persistent” researcher. 

“Her work focuses on efficient evolving and adaption of artificial intelligence algorithms to constantly changing environments for different users,” Hu said. “It’s really impressive and important work.” 

The winning paper aims to enable on-device learning on resource-limited field programmable gate arrays (FPGA). The team proposes an EF-Train, an efficient deep neural network training accelerator that enables FPGAs to continuously learn on the device – making it possible for current FPGA-based edge-level applications to achieve domain adaptation and personalization. (DOI: 10.1145/3505633)