Biography
I am a third-year postgraduate student of the Institute of Image Communication and Networks Engineering in theDepartment of Electronic Engineering(EE), Shanghai Jiao Tong University(SJTU), supervised by Prof. Shibao Zheng. I completed my bachelor degree in Electronic Engineering at SJTU. I was selected into the ”Excellent Engineer Training Program” in my sophomore year. And in 2019, I received the postgraduate recommendation from the Department of EE, SJTU. From 2018 to 2019, I was an research intern in Intel Asia Pacific R&D Center. During this period, I finished my graduation project and won the ”Excellent Intern” prize of 2019.
From March 2020 until now, I have been studying
in the TSAIL Group of State Key Lab. of Intelligent Tech. & Sys., Department of Computer Science and Technology, Tsinghua University, as a visiting student, supervised by Prof. Jun Zhu/Hang Su. And I'm now also an intern in HUAWEI Noah’s Ark Lab, working on the OOD generalization of object detection.
My research areas include machine learning, deep learning and their applications in computer vision. At present, I focus on
the adversarial robustness of deep learning.
Education
Shanghai Jiao Tong University
Shanghai, China
B.S. in Information Engineering
Sep. 2015 ‑ Jun. 2019
M.S. in Electronic and Communication Engineering Sep. 2019 ‑ Mar. 2022
Competitions
-
2021, 2nd Place, TianChi/CVPR2021 AIC Phase VI Track2: Unrestricted Adversarial Attacks on ImageNet.
-
2020, 8th Place, TianChi/CIKM2020 AIC Phase IV: Adversarial Challenge on Object Detection.
-
2019, 2nd Prize of Global Finals (top3), Huawei ICT Innovation Competition.
-
2019, Winning Award, Huawei Developer Competition IoT Track.
-
2018, National 1st Prize/Huawei Innovation Award, National IoT Contest.
-
2018, H Prize, Mathematical Contest In Modeling.
Publications
-
Improving Visual Quality of Unrestricted Adversarial Examples with Wavelet‑VAE
Wenzhao Xiang, Chang Liu, and Shibao Zheng.
ICML 2021 Workshop on Adversarial Machine Learning.
-
Improving Robustness of Adversarial Attacks Using an Affine-invariant Gradient Estimator
Wenzhao Xiang*, Hang Su*, Chang Liu, Yandong Guo, and Shibao Zheng.
arXivpreprint arXiv:2109.05820, Under Review of IEEE Transactions on Information Forensics and Security(TIFS).
-
Improving Model Generalization by On‑manifold AdversarialAugmentation in the Frequency Domain”
Chang Liu*, Wenzhao Xiang*, Hang Su, Yuan He, Hui Xue, Shibao Zheng.
Under Review of Internaltional Conference on Computer Vision and Pattern Recogintion 2022(CVPR 2022).
-
Model-Agnostic Meta-Attack: Towards Reliable Evaluation of
Adversarial Robustness
Xiao Yang, Yinpeng Dong, Wenzhao Xiang, Tianyu Pang, Hang Su, Jun Zhu.
Under Review of International Conference on Learning Representations 2022 (ICLR 2022).
Patents
Projects
-
2020, TSAIL Group, the Development of "Adversarial Robustness Evaluation for Safety" (ARES) Platform.
-
2020, Institute of Image Communication and Networks Engineering in SJTU, Grading Diagnosis of Human
Cataract Based on Slit-lamp Image.
-
2019, Google Summer of Code 2019, the Optimization of OpenCV.js with WebAssmbly Threads and SIMD.
-
2019, Intel Asia Pacific R&D Center, Implementation and Optimization of Web Machine Learning Inference
Engine Based on WebAssembly (graduation project of B.S.).
-
2018, Ministry of Education&Huawei, Design of Smart Home Based on OceanConnect Platform.
-
2018, Excellent Engineer Training Program, Design of Intelligent Tracking and Obstacle Avoidance Remote
Control Car Based on Arduino and Android Platform.
Honors & Awards
- Intel Scholarship, 2020.
- First Grade of Outstanding Postgraduate Scholarship of SJTU, 2020.
- Excellent Intern of Intel Asia Pacific R&D Center, 2019.
- C Scholarship of SJTU, 2018.
- National Encouragement Scholarship, 2017.
- Cyrus Tang Scholarship, 2016.
Last update: December 2021 by Wenzhao Xiang