About

I am a fourth-year Ph.D. candidate from Department of Computer Science and Engineering, University of Notre Dame. My research interest includes efficient and robust machine leanrning systems, efficient LLMs inference and deployment and optimization model under system constraint. Before my PhD study, I received bachelor and master degree from Huazhong University of Science and Technology.

You can find my CV here.

I am looking for industry internship, email: yqin3 [at] nd [dot] edu.

News

• [02/2025] Our work on uncertainty-aware and energy-efficient ML inference was accepted to ISSCC 2025.
  Uncertainty-aware ML inference system for real-time ventricular arrhythmia detection under ultra-low energy constraints.

• [01/2025] Our paper on negative feedback–based training for robust and efficient neural networks was accepted to IEEE TCAD.
  NeFT: Negative Feedback Training to Improve Robustness of Neural Networks under noisy and constrained inference.

• [12/2024] Our work was selected as a Best Paper Award Candidate at ICCAD 2024.
  Uncertainty-aware learning and inference for biomedical ML systems.

• [10/2024] Two papers on efficient and robust ML inference were accepted to ICCAD 2024.
  Topics include parameter sharing for efficient deployment and uncertainty-aware biomedical ML systems.

• [09/2024] Our work on efficient CNN inference for real-time healthcare applications was accepted to ASP-DAC 2025.

• [11/2023] Our paper on worst-case robust training under noisy inference conditions received the Best Paper Award at ICCAD 2023.

• [2023] Our survey on memristive neural networks for edge intelligence was published in Advanced Intelligent Systems and featured as a Back Cover Article.

Talks

• Department of Computer Science, Shandong University (SDU), Aug 2024
  Efficient and Robust Machine Learning under System and Noise Constraints

• Department of Electrical Engineering, Zhejiang University (ZJU), Aug 2024
  Uncertainty-aware ML Inference for Real-World Deployment

• University of Michigan – Shanghai Jiao Tong University Joint Institute, Shanghai Jiao Tong University, Aug 2024
  Robust and Efficient ML Systems: From Training to Deployment

• Department of Electrical Engineering, Southern University of Science and Technology (SUSTech), Jul 2024
  Efficient ML Inference for Resource-Constrained Systems

• AI Chip Center for Emerging Smart Systems (ACCESS), Hong Kong University of Science and Technology, Jun 2024
  End-to-End Efficient ML Systems under Practical Constraints

Research Experiences

• Research Assistant
  University of Notre Dame, Notre Dame, IN | Aug 2022 - present
  • Developed robust and efficient training and inference methods for machine learning models, including large language models (LLMs), by introducing a negative feedback–based training strategy and a lightweight parameter-sharing mechanism to reduce inference cost while maintaining accuracy under noise and system constraints.
  • Analyzed prediction uncertainty to characterize model reliability under noisy and constrained inference conditions, enabling robust and risk-aware deployment.
  • Published multiple first-author papers in top-tier conferences and journals on efficient and robust machine learning.
• Research Internship
  AI Chip Center for Emerging Smart Systems (ACCESS), Hong Kong | May 2024 - July 2024
  • Developed an efficient ML pipeline for real-time ventricular arrhythmia detection under strict latency and energy constraints.
  • Applied quantization and pruning techniques to reduce inference cost while preserving detection accuracy in a deployment-oriented setting.
  • Validated the approach in a realistic end-to-end system, including real-time inference and monitoring, demonstrating reliable performance for healthcare applications.
• Research Assistant
  Huazhong University of Science and Technology, Wuhan, China | Aug 2018 - Jun 2022
  • Conducted research on low-bit and quantization-aware learning methods for CNNs, improving robustness and accuracy-efficiency trade-offs under non-ideal inference conditions.
  • Published two papers in journals, including one featured as a journal back-cover article.

Publications

Journal

1. Yifan Qin⭐, Zheyu Yan, Dailin Gan, Jun Xia, Zixuan Pan, Wujie Wen, Xiaobo Sharon Hu, and Yiyu Shi. “NeFT: Negative Feedback Training to Improve Robustness of Compute-In-Memory DNN Accelerators”. In: IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD’25).

2. Han Bao, Yifan Qin⭐, Jia Chen, Ling Yang, Jiancong Li, Houji Zhou, Yi Li, and Xiangshui Miao. “Quantization and sparsity-aware processing for energy-efficient NVM-based convolutional neural networks”. In: Frontiers in Electronics (FE’22).

3. Yifan Qin⭐, Han Bao, Feng Wang, Jia Chen, Yi Li, and Xiangshui Miao. “Recent progress on memristive convolutional neural networks for edge intelligence”. In: Advanced Intelligent Systems (AIS’20) (Journal Back Cover).

4. Yifan Qin⭐, Rui Kuang, Xiaodi Huang, Yi Li, Jia Chen, and Xiangshui Miao. “Design of high robustness BNN inference accelerator based on binary memristors”. In: IEEE Transactions on Electron Devices (TED’20).

Conference

1. Zixuan Pan, Jun Xia, Zheyu Yan, Guoyue Xu, Yifan Qin⭐, Xueyang Li, Yawen Wu, Zhenge Jia, Jianxu Chen, Yiyu Shi, “Rethinking Medical Anomaly Detection in Brain MRI: An Image Quality Assessment Perspective,” in Proc. of IEEE International Conference on Bioinformatics and Biomedicine (BIBM’25).

2. Yifan Qin⭐, Zhenge Jia, Zheyu Yan, Jay Mok, Manto Yung, Yu Liu, Xuejiao Liu, Wujie Wen, Luhong Liang, Kwang-Ting Tim Cheng, X. Sharon Hu and Yiyu Shi, “A 10.60 μW 150 GOPS Mixed-Bit-Width Sparse CNN Accelerator for Life-Threatening Ventricular Arrhythmia Detection,” in Proc. of the Asia and South Pacific Design Automation Conference (ASP-DAC’25).

3. Jianbo Liu, Zephan Enciso, Boyang Cheng, Likai Pei, Steven Davis, Yifan Qin⭐, Zhenge Jia, Xiaobo Sharon Hu, Yiyu Shi and Ningyuan Cao, “A 65nm Uncertainty-quantifiable Ventricular Arrhythmia Detection Engine with 1.75μJ per Inference,” in Proc. of IEEE International Solid- State Circuits Conference (ISSCC’25).

4. Yifan Qin⭐, Zheyu Yan, Wujie Wen, Xiaobo Sharon Hu, and Yiyu Shi, “Sustainable Deployment of Deep Neural Networks on Non-Volatile Compute-in-Memory Accelerators”. In: International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS’24).

5. Likai Pei*, Yifan Qin*⭐, Zephan M. Enciso, Boyang Cheng, Jianbo Liu, Steven Davis, Zhenge Jia, Michael Niemier, Yiyu Shi, X. Sharon Hu and Ningyuan Cao. “Towards Uncertainty-Quantifiable Biomedical Intelligence: Mixed-signal Compute-in-Entropy for Bayesian Neural Networks”. In: IEEE/ACM International Conference on Computer-Aided Design (ICCAD’24). (*Equal contribution)(acceptance rate 24%)(2024 William J. McCalla Best Paper Award Candidate)(10 out of 802 submissions)

6. Yifan Qin⭐, Zheyu Yan, Zixuan Pan, Wujie Wen, Xiaobo Sharon Hu, and Yiyu Shi. “TSB: Tiny Shared Block for Efficient DNN Deployment on NVCIM Accelerators”. In: IEEE/ACM International Conference on Computer-Aided Design (ICCAD’24).(acceptance rate 24%)

7. Zheyu Yan, Yifan Qin⭐, Xiaobo Sharon Hu, and Yiyu Shi. “On the viability of using LLMs for SW/HW co-design: An example in designing CiM DNN accelerators”. In: 2023 IEEE 36th International System-on-Chip Conference (SOCC’23).

8. Zheyu Yan, Yifan Qin⭐, Wujie Wen, Xiaobo Sharon Hu, and Yiyu Shi. “Improving realistic worst-case perfor- mance of NVCiM DNN accelerators through training with right-censored gaussian noise”. In: 2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD’23). (2023 William J. McCalla Best Paper Award)(2 out of 750 submissions)

Patent

1. A hardware neural network batch normalization system
   {CN202011251999.9 · Issued May 20, 2022} Yi Li, Yifan Qin, Xiangshui Miao
2. A matrix-vector multiplication circuit and calculation method
   {CN201910792384.8 · Issued Oct 8, 2021} Yi Li, Yifan Qin, Xiangshui Miao

Review Experience

• Scientific Reports
• Great Lakes Symposium on VLSI Conference (GLSVLSI’25)
• the International Conference on Acoustics, Speech, and Signal Processing (ICASSP’25)

Efficient and Robust ML Inference: End-to-End System Demo

• 06/2024 Real-Time Ventricular Arrhythmia Detection: Efficient ML Inference System — ACCESS Center, HK

  Developed an end-to-end machine learning inference system for life-threatening ventricular arrhythmia detection under strict latency and energy constraints.
  The system demonstrates how efficient and robust ML inference can be achieved in realistic deployment environments.

  Key techniques include:

  • Quantization- and pruning-aware model design to reduce inference cost while preserving detection accuracy
  • System-aware inference optimization to meet real-time latency and energy budgets
  • End-to-end validation with real-time monitoring, illustrating reliable deployment of ML models in safety-critical settings

  Demo snapshots: real-time inference pipeline, hardware-integrated deployment, and monitoring interface.

  

• 02/2025 Uncertainty-Aware ML Inference for Real-Time Ventricular Arrhythmia Detection — ISSCC, San Francisco

  Developed an uncertainty-aware machine learning inference system for life-threatening ventricular arrhythmia detection under strict energy and reliability constraints.

  The system integrates Bayesian neural network inference with uncertainty quantification to enable reliable decision-making under:

  • Out-of-distribution (OOD) input data
  • Hardware non-idealities variations
  • Ultra-low energy budgets (1.75 μJ per inference)

  I led the design of the ML algorithms and uncertainty modeling, and contributed to end-to-end system integration and real-time monitoring, demonstrating robust deployment of uncertainty-aware ML inference in safety-critical environments.

  Demo snapshots: real-time uncertainty-aware inference pipeline and system validation. <!–

  

  –>

Honors & Awards

• William J. McCalla Best Paper Award, IEEE/ACM International Conference on Computer-Aided Design (ICCAD), 2023
  (2 out of 750 submissions)

• William J. McCalla Best Paper Award Candidate, IEEE/ACM ICCAD, 2024
  (10 out of 802 submissions)

• Young Fellow, Design Automation Conference (DAC), 2023–2025
  (selected for three consecutive years)

• Outstanding Graduate, Huazhong University of Science and Technology (HUST), 2020–2021

• National Second Prize, China Undergraduate Mathematical Modeling Contest, 2015

About Me (Optional)

• President
  Table Tennis Association of Huazhong University of Science and Technology
  Wuhan | Aug 2014 - Aug 2015
  Manage and organize school sports competitions and liaise with the Sports Academy secretary and finance department for project reporting and financial reimbursement

• Member
  Team of Undergraduate Mathematical Modeling of Huazhong University of Science and Technology
  Wuhan | Mar 2014 - Sep 2014
  National 2nd Prize in China Contemporary Undergraduate Mathematical Contest in Modeling

• Athlete
  College team of table tennis in Huazhong university of science and technology
  Wuhan | May 2014 - Sep 2016
  It is an honor to serve as one of the college team with many professional players, many of whom have become my good friends and have tought me to stock trading :)

• Museum Docent
  Wuhan Museum | Aug 2015 - Aug 2016

  • The museum’s opening introduction. (PIC: Wuhan museum, main hall painting)

  

  • The Ming dynasty (1368 AD to 1644 AD) and Qing dynasty (1644 AD to 1911 AD) artifacts historical introduction. (PIC: Artifact example)

  

Outside of research, I have a long-standing interest in art and creative expression.
One of my paintings was included in a provincial-level art textbook published by the People’s Education Press in China.

I enjoy drawing and writing, activities that foster imagination and long-term creative thinking.
I also stay active through sports, including table tennis and soccer, and regularly work out at the gym.

“风筝不断线” — Guanzhong Wu