Skip to content
Avatar
🐈
🐈

Highlights

  • Pro
Block or Report

Block or report vtu81

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
vtu81/README.md

Hi there 👋

visitors

Hello, I am Tinghao Xie 谢廷浩, a first year ECE PhD student at Princeton. I just received my Bachelor degree from Computer Science and Technology at Zhejiang University.

I found the robustness of machine learning being both a “dark cloud” and an attractive perspective to work on. Specifically, my research interest could be described in two aspects: First, I intend to study and solve security concerns involving current non-robust deep learning models; Second, I would like to better understand AI’s behaviors and make their predictions more human-like through explainable and causal methods. In summary, I hope to fully explore the breadth and depth of secure, robust, and reliable AI. Yet, I'm always on my way looking for things that intrigue me, and tend to hold an open mind for whatever is coming. I (wish to) have fun doing research.

Earlier, I finished our project Subnet Replacement Attack (SRA)[paper][code] (accepted by CVPR'22) of the first gray-box and physically realizable backdoor weight attack, collaborating with Xiangyu Qi at Princeton University, advised by Principal Researcher Jifeng Zhu at Tencent Zhuque Lab and Prof. Kai Bu at ZJU. I was also fortunate to work with Prof. Ting Wang on backdoor certification[blog] and backdoor restoration[blog] at Pennsylvania State University as an intern, meanwhile co-advised by Prof. Shouling Ji at ZJU NESA Lab. Even before, I worked with Prof. Jianhai Chen, designed and implemented Enchecap[code], an encrypted (enclave-based) heterogeneous calculation protocol.

News & Facts

vtu81's GitHub stats Top Langs

Pinned

  1. A compact toolbox for backdoor attacks and defenses.

    Python 10

  2. Official implementation of (CVPR 2022 Oral) Towards Practical Deployment-Stage Backdoor Attack on Deep Neural Networks.

    Jupyter Notebook 6 2

  3. Applying backdoor attacks to BadNet on MNIST and ResNet on CIFAR10.

    Jupyter Notebook 6

  4. Enchecap Public

    An encrypted (enclave-based) heterogeneous calculation protocol based on Nvidia CUDA and Intel SGX.

    C 5

  5. WebGL basic engine and Tron demo.

    JavaScript 4

  6. Forked from furry-potato-maker/ZJU-Clock-In

    探究浙江大学健康打卡的原理与对抗策略

    Python 75 242

454 contributions in the last year

Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Mon Wed Fri

Contribution activity

August 2022

3 contributions in private repositories Aug 10 – Aug 16

Seeing something unexpected? Take a look at the GitHub profile guide.