Shaoyi Huang
Shaoyi is a 4-th year Ph.D. student in the School of Computing at University of Connecticut,
working with Prof. Caiwen Ding and Prof. Omer Khan as a member of Intelligent & Efficient Systems Laboratory.
Her research agenda is grounded in advancing AI systems, including algorithm-system co-design for AI acceleration, emerging deep learning models inference acceleration (i.e., Transformer and LLM), privacy preserving machine learning, and machine learning for EDA.
Shaoyi’s work has been published in high-impact conferences such as HPCA, ASPLOS, SC, DAC, ICCAD, ACL, ICCV, NeurIPS, IJCAI, etc.
Her work on FPGA-based language models acceleration through sparse attention and dynamic pipelining won DAC'22 Publicity Paper Award. She is the recipient of Cigna Graduate Fellowship in 2021, Eversource Energy Graduate Fellowship in 2022, and Synchrony Fellowship in 2022.
She was awarded both Predoctoral Prize for Research Excellence and GE Fellowship for Excellence from UConn in 2022 and 2023.
She was invited to serve as a student panelist to share experience with the female researchers and K-12 students at the 6th Workshop for Women in Hardware and Systems Security (WISE).
Shaoyi is on the 2024 academic job market, please do not hesitate to reach out for any relevant job openings.
Email  / 
CV  / 
Google Scholar  / 
Github
|
|
Research Interests
Sparsity & Parallelism Exploitation Sparse Training
[HPCA'24],
[ASPLOS'24],
[DAC'23(a)],
[DAC'23(b)],
[ICCD'22]
Emerging Deep Learning Models Inference Acceleration
[ICCAD'23],
[ACL'22],
[DAC'22],
[ICCD'22],
[SC'21],
[ICCAD'21],
[ISQED'21],
[GLSVLSI'21],
[DAC'24(a) submitted],
[DAC'24(b) submitted]
Privacy Preserving Machine Learning
[ICCV'23],
[NeurIPS'23],
[IEEE BigData'23],
[AAAI'24 submitted]
Machine Learning for EDA
[ASPLOS'24 submitted]
|
News
- 10/2023: One paper was accepted to ASPLOS 2024.
- 10/2023: One paper was accepted to HPCA 2024.
- 10/2023: Received travel grant from CACC. Thanks, CACC!
- 09/2023: Honored to be invited to be the student panelist and received student travel grant from WISE 2023.
- 09/2023: One proposal on Efficient Multi-party Computation-based Private Inference was accepted to WISE 2023 (The 6th Workshop for Women in Hardware and Systems Security).
- 09/2023: One paper was accepted to NeurIPS 2023.
- 09/2023: Honored to be invited to give a guest lecture at the University of Rochester.
- 09/2023: Invited to serve as a PC member for SDM 2024.
- 09/2023: Honored to present at TechCon organized by Semiconductor Research Corporation (SRC). See u in Austin.
- 07/2023: One paper was accepted to ICCAD 2023.
- 07/2023: One paper was accepted to ICCV 2023.
- 07/2023: Invited to serve as a PC member for AAAI 2024.
- 06/2023: Received the GE Fellowship for Excellence from UConn School of Engineering.
- 05/2023: Honored to give a talk and a poster presentation at Semiconductor Research Corporation (SRC) AIHW & CADT Annual Review in IBM Research. See u in San Jose.
- 05/2023: Received the Predoctoral Prize for Research Excellence from UConn CSE.
- 03/2023: Honored to be invited to give a talk at MLNLP.
- 03/2023: One paper was accepted to IJCAI 2023.
- 03/2023: Two papers on sparse training were accepted to DAC 2023. See u in San Francisco.
- 01/2023: Invited to serve as a reviewer for the First Workshop on DL-Hardware Co-Design for AI Acceleration at AAAI 2023.
- 12/2022: Invited to serve as a PC member for KDD 2023.
- 08/2022: Received the GE Fellowship for Excellence from the School of Engineering. Thanks for the support!
- 07/2022: Invited to serve as a PC member for AAAI 2023.
- 07/2022: Received the Synchrony Fellowship from Connecticut Advanced Computing Center. Thanks, CACC!
- 05/2022: Started to work at TikTok (ByteDance) as a research intern.
- 05/2022: Received the Predoctoral Prize for Research Excellence from UConn CSE.
- 05/2022: Received the Eversource Energy Graduate Fellowship from Eversource Energy Center. Thanks, Eversource!
- 04/2022: Our DAC 22 paper “A Length Adaptive Algorithm-Hardware Co-design of Transformer on FPGA Through Sparse Attention and Dynamic Pipelining” was recognized as a Publicity Paper.
- 02/2022: One paper was accepted to ACL 2022.
- 02/2022: One paper was accepted to DAC 2022.
- 02/2022: One paper was accepted to ISQED 2022.
- 08/2021: Received the Cigna Graduate Fellowship from UConn CSE. Thanks, Cigna!
- 07/2021: One paper was accepted to ICCAD 2021.
- 06/2021: One paper was accepted to SC 2021.
- 04/2021: Three papers were accepted to GLSVLSI 2021.
- 02/2021: One paper was accepted to ISQED 2021.
|
|
MaxK-GNN: Towards Theoretical Speed Limits for Accelerating Graph Neural Networks Training
H. Peng, X. Xie, K. Shivdikar, M. A. Hasan, J. Zhao, S. Huang, O. Khan, D. Kaeli, C. Ding
[ASPLOS 2024]
Paper
|
|
PruneGNN: An Optimized Algorithm-Hardware Framework for Graph Neural Network Pruning
D. Gurevin, M. Shan, S. Huang, M. A. Hasan, C. Ding, O. Khan
[HPCA 2024]
Paper
(Acceptance rate: 75/410=18.3%)
|
|
Dynamic Sparse Training via Balancing the Exploration-Exploitation Trade-off
S. Huang, B. Lei, D. Xu, H. Peng, Y. Sun, M. Xie, C. Ding
[DAC 2023]
Paper
(Acceptance rate: 263/1156=22.7%)
|
|
Neurogenesis Dynamics-inspired Spiking Neural Network Training Acceleration
S. Huang, H. Fang, K. Mahmood, B. Lei, N. Xu, B. Lei, Y. Sun, D. Xu, W. Wen, C. Ding
[DAC 2023]
Paper
(Acceptance rate: 263/1156=22.7%)
|
|
AutoReP: Automatic ReLU Replacement for Fast Private Network Inference
S. Huang*, H. Peng*, T. Zhou*, W. Wen, X. Xu, C. Ding, et al.
[ICCV 2023]
Paper / Code
(Acceptance rate: 26.8%)
|
|
Towards Lossless Head Pruning through Automatic Peer Distillation for Large Language Models
B. Li, Z. Wang, S. Huang, M. Bragin, J. Li, C. Ding
[IJCAI 2023]
Paper
(Acceptance rate: 15%)
|
|
Accel-GCN: High-Performance GPU Accelerator Design for Graph Convolution Networks
X. Xie, H. Peng, A. Hasan, S. Huang, J. Zhao, H. Fang, W. Zhang, T. Geng, O. Khan, C. Ding
[ICCAD 2023]
Paper / Code
|
|
LinGCN: Structural Linearized Graph Convolutional Network for Homomorphically Encrypted Inference
H. Peng, R. Ran, Y. Luo, J. Zhao, S. Huang, K. Thorat, T. Geng, C. Wang, X.lin Xu, W. Wen, C. Ding
[NeurIPS 2023]
Paper / Code
(Acceptance rate: 26.1%)
|
|
Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm
S. Huang, D. Xu, I. E. Yen, S. Chang, B. Li, C. Ding, et al.
[ACL 2022]
Paper / Code
(Acceptance rate: 714/3350=21.3%)
In collaboration with Moffett AI
|
|
A Length Adaptive Algorithm-Hardware Co-design of Transformer on FPGA Through Sparse Attention and Dynamic Pipelining
S. Huang*, H. Peng*, S. Chen, B. Li, T. Geng, A, Li, W. Jiang, W, Wen, J, Bi, H. Liu, C. Ding
[DAC 2022]
Paper
Publicity paper award
|
|
Towards Sparsification of Graph Neural Networks
H. Peng, D. Gurevin, S. Huang, T. Geng, W. Jiang, O. Khan, C. Ding
[ICCD 2022]
Paper / Code
|
|
Analyzing and Defending against Membership Inference Attacks in Natural Language Processing Classification
Y. Wang, N. Xu, S. Huang, K. Mahmood, D. Guo, C. Ding, W. Wen, S. Rajasekaran
[IEEE BigData 2022]
Paper
|
|
CoDG-ReRAM: An Algorithm-Hardware Co-design to Accelerate Semi-Structured GNNs on ReRAM
Y. Luo, P. Behnam, K. Thorat, Z. Liu, H. Peng, S. Huang, S. Zhou, O. Khan, A. Tumanov, C. Ding, T. Geng
[ICCD 2022]
Paper
|
|
E.T.: Re-Thinking Self-Attention for Transformer Models on GPU
S. Huang*, S. Chen*, S. Pandey, B. Li, G. Gao, C. Ding, H. Liu
[SC 2021]
Paper / Code
(Acceptance rate: 86/365=23.6%)
|
|
Accelerating Framework of Transformer by Hardware Design and Model Compression Co-Optimization
P. Qi, E. Sha, Q. Zhuge, H. Peng, S. Huang, Z. Kong, Y. Song, B. Li
[ICCAD 2021]
Paper
|
|
HMC-Tran: A Tensor-core Inspired Hierarchical Model Compression for Transformer-based DNNs on GPU
S. Huang, S. Chen, H. Peng, D. Manu, Z. Kong, G. Yuan, L. Yang, S. Wang, H. Liu, C. Ding
[GLSVLSI 2021]
Paper
|
|
Accelerating Transformer-based Deep Learning Models on FPGAs Using Column Balanced Block Pruning
H. Peng, S. Huang, T. Geng, A. Li, W. Jiang, H. Liu, W. Wang, C. Ding
[ISQED 2021]
Paper
|
Teaching
- Guest Lecturer
- Teaching Assistant at University of Connecticut
CSE 4502/5717 - BigData Analytics, Spring 2023
Instructor: Prof. Suining He   
CSE5819 - Introduction to Machine Learning, Fall 2022
Instructor: Prof. Fei Miao   
As a co-instructor, I designed final project, developed and delivered coding tutorials, led final project presentation courses, supervised students on course projects, held office hours and graded assignments.
- Teaching Assistant at University of Rochester
Microcontroller, Spring 2018
Instructor: Prof. Qiang Lin   
Circuits & Signals LAB, Fall 2017
Instructor: Prof. Jack G. Mottley   
I led the laboratory sessions and provided hands-on instruction to around 100 undergraduate students.
- Mentering
Amit Hasan, Ph.D. Student at University of Connecticut
Topic: LLM Inference Acceleration
Yifan Shan, Undergraduate at University of Connecticut (Now master at Cornell Tech)
Project: Evaluating the Impact of Preferential Trade Agreements on Agricultural and Food Trade: New Insights from Natural Language Processing and Machine Learning
Jiwon Kim, Undergraduate at University of Connecticut
Project (NSF-REU): Utilization of DeepShift for Privacy Based Machine Learning
Alison Menezes, Undergraduate at Clemson University
Project (NSF-REU): Deep Leakage from Gradients on GNNs
Maryam Abuissa, Undergraduate at Amherst College
Project (NSF-REU): Sequestered Encryption for GPU
|
Internship
May 2022 - Aug 2022, Research Intern, TikTok, Austin, TX
May 2021 - Aug 2021, Research Intern, Moffett AI, Los Altos, CA
|
Talks
Student panelist - The 6th Workshop for Women in Hardware and Systems Security
Oct 2023, California State University, Fullerton, Fullerton, CA
Towards Efficient Model Inference and Training
Sep 2023, University of Rochester
Guest Lecture of ECE 403-1: Advanced Computer Architecture for Machine Learning.
Host: Prof. Tony Geng
Towards Efficient Training and Inference Under Pretrain-and-Finetune Paradigm
Sep 2023, TechCon - Semiconductor Research Corporation (SRC), Austin, TX
Exploring Extreme Sparsity in Training and Inference for Graph Neural Networks to Achieve High Performance
Scaling on Large Core Count Machines
May 2023, Semiconductor Research Corporation (SRC) AIHW \& CADT Annual Review, IBM Research, San Jose, CA
Towards Efficient Model Inference and Training
Apr 2023, Machine Learning and Natural Language Processing Community (MLNLP)
MLNLP Outstanding Speaker
|
Honors and Awards
-
NeurIPS Travel Grant from CACC, 2023
-
WISE Student Travel Award, 2023
-
GE Fellowship of Excellence, 2023
-
Predoctoral Prize for Research Excellence, 2023
-
MLNLP Outstanding Speaker, 2023
-
GE Fellowship for Excellence, 2022
-
Synchrony Fellowship, 2022
-
DAC Publicity Paper Award, 2022
-
Predoctoral Prize for Research Excellence, 2022
-
Eversource Energy Graduate Fellowship, 2022
-
DAC Young Fellow, 2021
-
Cigna Graduate Fellowship, 2021
|
Professional Services
-
Panelist
-
The 6th Workshop for Women in Hardware and Systems Security (WISE)
-
Program Committee Member
-
ACL'24
-
SDM'24
-
AAAI'24
-
NeurIPS'23 Datasets and Benchmarks
-
KDD'24, 23
-
AAAI'23
-
DCAA'23
-
Conference External Reviewer
-
NeurIPS'23
-
DAC'23, 22, 21
-
ICCAD'23, 22, 21
-
AAAI'23, 22, 21
-
EMNLP'23, 22
-
IJCAI'23, 22, 21
-
ICCD'23, 22
-
EACL'22
-
FPGA'21
-
NAACL'21
-
Journal Reviewer
-
Neurocomputing
-
Pattern Recognition
-
Engineering Applications of Artificial Intelligence
-
Neural Networks
-
Journal of Systems Architecture
-
IEEE Transactions on Neural Networks and Learning Systems
-
Journal External Reviewer
-
ACM Transactions on Architecture and Code Optimization
-
IEEE Transactions on Neural Networks and Learning Systems
-
IEEE Network Magazine
-
Journal on Emerging Technologies in Computing Systems
|
|