Zecheng He


Ph.D Candidate

Department of Electrical Engineering, Princeton University

Email: zechengh@princeton.edu


I am a final-year PhD student in PALMS group, Department of Electrical Engineering at Princeton University, advised by Prof. Ruby B. Lee. I received M.A. in Electrical Engineering from Princeton University in 2017 and B.E. in Electronic Information Engineering from University of Science and Technology of China (USTC) in 2015, respectively.

Research



Selected Publications


  • Attacking and Protecting Data Privacy in Edge-Cloud Collaborative Inference Systems
    Zecheng He, Tianwei Zhang and Ruby B. Lee
    IEEE Internet of Things Journal, 2020, accepted, to appear (Journal, IF=9.515)

  • Miss the Point: Targeted Adversarial Attack on Multiple Landmark Detection
    Qingsong Yao, Zecheng He, Hu Han and S. Kevin Zhou
    MICCAI 2020, accepted, to appear

  • Model Inversion Attacks against Collaborative Inference
    Zecheng He, Tianwei Zhang and Ruby B. Lee
    Annual Computer Security Applications Conference (ACSAC'19) (accept rate 22%)

  • Sensitive-Sample Fingerprinting of Deep Neural Networks
    Zecheng He, Tianwei Zhang and Ruby B. Lee
    IEEE Conference on Computer Vision and Pattern Recognition (CVPR'19), 2019

  • Power-Grid Controller Anomaly Detection with Enhanced Temporal Deep Learning
    Zecheng He, Aswin Raghavan, Guangyuan Hu, Sek Chai and Ruby B. Lee
    TrustCom, 2019

  • How Secure Is Your Cache Against Side-channel Attacks?
    Zecheng He, and Ruby B. Lee
    IEEE/ACM International Symposium on Microarchitecture (MICRO'17), 2017 (accept rate 18%)

  • Cross-Scale Color Image Restoration Under High Density Salt-and-Pepper Noise
    Zecheng He, Ketan Tang and Lu Fang
    IEEE International Conference on Image Processing (ICIP'17), 2017

  • Machine Learning Based DDoS Attack Detection from Source Side in Cloud
    Zecheng He, Tianwei Zhang, and Ruby B. Lee
    IEEE International Conference on Cyber Security and Cloud Computing, 2017

    Preprint

  • New Models for Understanding and Reasoning about Speculative Execution Attacks
    Zecheng He, Guangyuan Hu, and Ruby B. Lee
    [arXiv]

  • Smartphone Impostor Detection with Built-in Sensors and Deep Learning
    Guangyuan Hu, Zecheng He and Ruby B. Lee
    [arXiv]

  • VerIDeep: Verifying Integrity of Deep Neural Networks through Sensitive-Sample Fingerprinting
    Zecheng He, Tianwei Zhang and Ruby B. Lee
    [arXiv]

  • Privacy-preserving Machine Learning through Data Obfuscation
    Tianwei Zhang, Zecheng He and Ruby B. Lee
    [arXiv]

Work Experience


    Internship

  • Research Intern, Google
    Multimodal dialogue team. May 2020 - Aug 2020, Mountain View, CA
    Multimodal UI embedding.

  • Software Engineer Intern, Facebook
    Machine learning track, Core ML team in Business Integrity. May 2019 - Aug 2019, Menlo Park, CA
    Detect policy-violating ads through machine learning.
    Evaluate the proposed models on BI top-level metrics. Work has been adopted across teams.

  • Research Intern, SRI International
    Jun 2017 - Sep 2017, Princeton, NJ
    Enhancing deep temporal model with statistical test for processor anomaly detection in power-grid systems.
    Real-time and high-reliable controller abnormal behavior detection system design.

  • Research Intern, Huawei Technologies R&D
    Jun 2016 - Sep 2016, Bridgewater, NJ
    Adaptive-depth convolutional neural network (CNN) for image style transfer.


Selected Awards


  • 1st place, Siemens Futuremaker Hackthon, 2018   [Press1] [Press2]
  • Gordon Y.S. Wu Fellowship in Engineering, 2015-2019
  • Guo Moruo Scholarship, USTC, 2015   (top 1%, highest award for undergraduate excellence)
  • Honor Degree, Elites Class in Information Science, USTC, 2015
  • National Scholarship, China, 2014 (top 2%)
  • Honorable Fellowship, Institute of Electronics China Academy of Science, 2013

Talks


  • "Security Meets Deep Learning"
    Princeton AI Seminar, Princeton, Apr 2019

  • "How Secure Is Your Cache Against Side-Channel Attacks?"
    SRC Techcon, Austin TX, Sep 2018

  • "Security in Deep Learning"
    Z2AI, Princeton NJ, Aug 2018

  • "Deep Learning Meets Security"
    SRI International, Princeton NJ, Jun 2018

  • "Modeling and Evaluatiing Cache Resilience Against Side-channel Attacks"
    Princeton Research Day, Princeton NJ, May 2018

  • "Security Verification of Resilience to Cache Side-Channel Attacks"
    SRC T3S Annual Meeting, Portland OR, Sep 2016


Professional Service


    Session Chair

  • Session 4, IEEE International Conference on Cyber Security and Cloud Computing, New York, NY, 2017

    Reviewer

  • 2020: ISCA, MICRO, IEEE TIM, IEEE Access
  • 2019: S&P, IEEE SPL
  • 2018: CCS, S&P, MICRO, HPCA
  • 2017: MICRO, HASP, IEEE ToC
  • 2016: MICRO

Teaching


  • Teaching Assistant, ELE 115 Introduction to Computing: Programming Autonomous Vehicles, Princeton University, Spring'20
  • Teaching Assistant, ELE 470 Smartphone Security and Architecture, Princeton University, Fall'17
  • Teaching Assistant, Signals and Systems, University of Science and Technology of China (USTC), Spring'15
  • Teaching Assistant, The C Language Programing, University of Science and Technology of China (USTC), Fall'14

Education


  • Princeton University
    Ph.D, Electrical Engineering, Sep 2015 -
    M.A., Electrical Engineering, Sep 2015 - May 2017
    Advisor: Prof. Ruby B. Lee
    GPA: 3.92/4.00

  • University of Science and Technology of China (USTC)
    B.E., Electronic Information Engineering, Sep 2011 - May 2015
    GPA 4.00/4.30, rank 1/252