Shuya Feng

Shuya Feng

Assistant Professor in Computer Science

Biography

I'm an Assistant Professor in the Department of Computer Science at the University of Alabama Birmingham (UAB). My research lies at the intersection of privacy, security, and machine learning, with a particular focus on developing privacy-preserving technologies for real-world applications. I'm especially interested in differential privacy mechanisms for streaming data and distributed machine learning systems.

I earned my Ph.D. in Computer Science and Engineering from the University of Connecticut, where I was advised by Dr. Yuan Hong. Before that, I received my M.S. and B.S. in Computer Science from Illinois Institute of Technology and Harbin Institute of Technology, respectively.

๐ŸŽ“ I'm recruiting PhD students!
I am recruiting Ph.D. students to join my team! If you're passionate about privacy-preserving technologies, differential privacy, or machine learning security, feel free to reach out to me at fengs [AT] uab [DOT] edu.

Interests

  • Security and Privacy
  • Data Privacy & Differential Privacy
  • Responsible AI
  • ML Privacy & Security
  • LLM Privacy & Security
  • Usable Security

Education

๐ŸŽ“ Ph.D. Computer Science & Engineering, 2025
University of Connecticut
๐ŸŽ“ M.S. Computer Science, 2017
Illinois Institute of Technology
๐ŸŽ“ B.S. Computer Science, 2015
Harbin Institute of Technology

News

[Aug. 2025] Position Started as Assistant Professor at University of Alabama Birmingham!
[Nov. 2024] Talk Guest Lecture at CSE5173 Deep Learning, University of Cincinnati.
[July 2024] Award Received Conference Participation Award by The Graduate School UConn.
[May 2024] Award Received The Marion and Frederick Buckman Engineering Fellowship.
[May 2024] Internship Completed Research Intern position at Amazon, NYC.
[Apr. 2024] Award Received Travel Conferenceship by IEEE S&P.

Conference Papers

Harmonizing Differential Privacy Mechanisms in Federated Learning: Ensuring Boosted Accuracy and Convergence

(CODASPY 2025)
Shuya Feng, Meisam Mohammady, Hanbin Hong, Shenao Yan, Binghui Wang, Ashish Kundu, Yuan Hong
This work addresses privacy challenges in federated learning by harmonizing differential privacy mechanisms to ensure both privacy protection and model accuracy convergence.

Delay-allowed Differentially Private Data Stream Release

(NDSS 2025)
Xiaochen Li, Zhan Qin, Kui Ren, Chen Gong, Shuya Feng, Yuan Hong, Tianhao Wang
Novel approach to differentially private data stream release that allows controlled delays to improve privacy-utility trade-offs in streaming applications.

RankFlow: A Multi-Role Collaborative Reranking Workflow Utilizing Large Language Models

(WWW 2025)
Can Jin, Hongwu Peng, Anxiang Zhang, Nuo Chen, Jiahui Zhao, Xi Xie, Kuangzheng Li, Shuya Feng, Kai Zhong, Caiwen Ding, Dimitris N. Metaxas
Multi-role collaborative framework for reranking using large language models to improve information retrieval and ranking performance.

DPI: Ensuring Strict Differential Privacy for Infinite Data Streaming

(IEEE S&P 2025)
Shuya Feng*, Meisam Mohammady*, Han Wang, Xiaochen Li, Zhan Qin, Yuan Hong
Addresses the challenge of ensuring differential privacy in infinite data streams with strict privacy guarantees. [Acceptance Rate: 14.9%, *Equal Contribution]

Enhanced body composition estimation from 3d body scans

(BIOINFORMATICS 2024)
Boyuan Feng, Yijiang Zheng, Ruting Cheng, Khashaya Vaziri, Shuya Feng, James Hahn
Novel method for estimating body composition from 3D body scans using advanced computational techniques.

Towards Accurate and Stronger Local Differential Privacy for Federated Learning with Staircase Randomized Response

(CODASPY 2024)
Matta Varun, Shuya Feng, Han Wang, Shamik Sural, Yuan Hong
Improved local differential privacy mechanisms for federated learning with enhanced accuracy and stronger privacy guarantees. [Acceptance Rate: 34/160=21.25%]

A Model-Agnostic Approach to Differentially Private Topic Mining

(ACM KDD 2022)
Han Wang*, Jayashree Sharma*, Shuya Feng, Kai Shu, Yuan Hong
Framework for privacy-preserving topic mining that works across different ML models. [Acceptance Rate: 254/1695=14.99%, *Equal Contribution]

Security Analysis of Block Withholding Attacks in Blockchain

(IEEE ICC 2021)
Shuya Feng, Jia He, Maggie Cheng
Analysis of security vulnerabilities in blockchain mining protocols and mitigation strategies for block withholding attacks.

Teaching

CS334/534 Networking
University of Alabama Birmingham
Aug 2025 - Dec 2025 โ€ข Instructor
CSE4400 Computer Security
University of Connecticut
Jan 2023 - May 2023 โ€ข Teaching Assistant
CSE1010 Introduction to Computing for Engineers
University of Connecticut
Aug 2022 - Dec 2022 โ€ข Teaching Assistant
CS425 Database Organization
Illinois Institute of Technology
Jan 2022 - May 2022 โ€ข Teaching Assistant
CS528 Data Privacy and Security
Illinois Institute of Technology
Aug 2021 - Dec 2021 โ€ข Teaching Assistant

Contact

๐Ÿ“ง
fengs@uab.edu
๐Ÿข
Department of Computer Science
University of Alabama Birmingham