ReDS Lab (reds-lab)

ReDS Lab

reds-lab

Geek Repo

Responsible Data Science Lab @ Virginia Tech | AI Security & Privacy & More

Twitter:@reds_lab_vt

Github PK Tool:Github PK Tool

ReDS Lab 's repositories

Narcissus

The official implementation of the CCS'23 paper, Narcissus clean-label backdoor attack -- only takes THREE images to poison a face recognition dataset in a clean-label way and achieves a 99.89% attack success rate.

Language:PythonLicense:MITStargazers:104Issues:2Issues:10

LAVA

This is an official repository for "LAVA: Data Valuation without Pre-Specified Learning Algorithms" (ICLR2023).

Language:PythonLicense:MITStargazers:44Issues:0Issues:2

CLIP-MIA

This is an official repository for Practical Membership Inference Attacks Against Large-Scale Multi-Modal Models: A Pilot Study (ICCV2023).

Language:Jupyter NotebookLicense:MITStargazers:20Issues:0Issues:2

Meta-Sift

The official implementation of USENIX Security'23 paper "Meta-Sift" -- Ten minutes or less to find a 1000-size or larger clean subset on poisoned dataset.

Language:PythonStargazers:18Issues:2Issues:0

ASSET

This repository is the official implementation of the paper "ASSET: Robust Backdoor Data Detection Across a Multiplicity of Deep Learning Paradigms." ASSET achieves state-of-the-art reliability in detecting poisoned samples in end-to-end supervised learning/ self-supervised learning/ transfer learning.

Language:PythonLicense:MITStargazers:17Issues:3Issues:2

Universal_Pert_Cert

This repo is the official implementation of the ICLR'23 paper "Towards Robustness Certification Against Universal Perturbations." We calculate the certified robustness against universal perturbations (UAP/ Backdoor) given a trained model.

Language:PythonLicense:MITStargazers:12Issues:3Issues:1

projektor

This is an official repository for "Performance Scaling via Optimal Transport: Enabling Data Selection from Partially Revealed Sources" (NeurIPS 2023).

Language:PythonLicense:MITStargazers:11Issues:0Issues:0

BEEAR

This is the official Gtihub repo for our paper: "BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Models".

2d-shapley

This is an official repository for "2D-Shapley: A Framework for Fragmented Data Valuation" (ICML2023).

Language:Jupyter NotebookLicense:MITStargazers:4Issues:0Issues:1

privmon

This is an official repository for PrivMon: A Stream-Based System for Real-Time Privacy Attack Detection for Machine Learning Models (RAID 2023)

Language:PythonLicense:MITStargazers:4Issues:0Issues:0
Language:Jupyter NotebookLicense:Apache-2.0Stargazers:3Issues:0Issues:0

Knowledge-Enriched-DMI

The official implementation of the ICCV 2021 paper, "Knowledge-Enriched Distributional Model Inversion Attacks."

Language:PythonLicense:MITStargazers:3Issues:0Issues:0

I-BAU

Official Implementation of the ICLR 2022 paper, ``Adversarial Unlearning of Backdoors via Implicit Hypergradient''

Language:Jupyter NotebookLicense:MITStargazers:2Issues:0Issues:0

Nash-Meta-Learning

Official implementation of "Fairness-Aware Meta-Learning via Nash Bargaining." We explore hypergradient conflicts in one-stage meta-learning and their impact on fairness. Our two-stage approach uses Nash bargaining to mitigate conflicts, enhancing fairness and model performance simultaneously.

Language:Jupyter NotebookStargazers:2Issues:0Issues:0
Language:HTMLStargazers:2Issues:0Issues:0

frequency-backdoor

The official implementation of the ICCV 2021 paper, "Rethinking the backdoor attacks' triggers: A frequency perspective."

Language:Jupyter NotebookLicense:MITStargazers:1Issues:0Issues:0

preference-learning-with-rationales

This is the public repository for Data-Centric Human Preference Optimization with Rationales.

Language:PythonLicense:Apache-2.0Stargazers:1Issues:0Issues:0

dataselection

Projektor Website

Language:JavaScriptLicense:MITStargazers:0Issues:0Issues:0

reds-lab.github.io

Homepage portfolio of Reds Projects

Language:TypeScriptLicense:MITStargazers:0Issues:0Issues:0
Language:PythonLicense:MITStargazers:0Issues:0Issues:0