vtu81 Goto Github PK

adversarial-attacks-pytorch

PyTorch implementation of adversarial attacks.

auto_lirpa

[NeurIPS 2020]auto_LiRPA: An Automatic Linear Relaxation based Perturbation Analysis Library for Neural Networks

backdoor-toolbox

A compact toolbox for backdoor attacks and defenses.

backdoor_attack

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

bgphijacking

cos598d-pruning

Assignments for COS598D: System and Machine Learning

cos598d-serverless

enchecap

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

endc

Ensemble of Narrow DNN Chains.

fight-poison-with-poison

Code repository for the paper --- Fight Poison with Poison: Detecting Backdoor Poison Samples via Decoupling Benign Correlations

githubgraduation-2022

Join the GitHub Graduation Yearbook and "walk the stage" on June 11.

hwms

HomeWork Management System, helping administrators, teachers and students manage courses and homework, implemented as a bash shell script and based on MySQL. -- 浙江大学Linux程序设计短学期实验2 “作业管理系统”

iotwebsite

An IoT platform website, consisting of a MQTT server receiving info from (simulated) devices, a MySQL server storing data, an Express Node.js backend, a React with Material UI frontend (using the template material-kit-react).

llm-recipe

Fine-tuning Llama-3, Gemma, Mistral (derived from Meta's llama-recipe)

machine-learning-at-oxford

My code for practicals (or projects/labs) of course Machine Learning at Oxford.

minisql

MiniSQL: A simple single-user SQL engine. 一个精简型单用户SQL引擎(DBMS)

mips-translator

A simple MIPS instructions translator.

myshell

A simplified Linux shell written in C. -- 浙江大学Linux程序设计短学期实验2

naivevqa

A Visual Question Answering model implemented in MindSpore and PyTorch. The model is a reimplementation of the paper *Show, Ask, Attend, and Answer: A Strong Baseline For Visual Question Answering*. It's our final project for course DL4NLP at ZJU.

piclib

Personal picture library for markdown.

spectre-defense

Defending Against Backdoor Attacks Using Robust Covariance Estimation

trojanzoo

TrojanZoo provides a universal pytorch platform to conduct security researches (especially backdoor attacks/defenses) of image classification in deep learning.

vtu81

zju-clock-in

探究浙江大学健康打卡的原理与对抗策略