Madry Lab

cifar10_challenge

A challenge to explore adversarial robustness of neural networks on CIFAR10.

constructed-datasets

Datasets for the paper "Adversarial Examples are not Bugs, They Are Features"

context-cite

Attribute (or cite) statements generated by LLMs back to in-context information.

datamodels-data

Data for "Datamodels: Predicting Predictions with Training Data"

mnist_challenge

A challenge to explore adversarial robustness of neural networks on MNIST.

modelcomponents

Decomposing and Editing Predictions by Modeling Model Computation

modeldiff

ModelDiff: A Framework for Comparing Learning Algorithms

photoguard

Raising the Cost of Malicious AI-Powered Image Editing

robustness

A library for experimenting with, training and evaluating neural networks, with a focus on adversarial robustness.

trak

A fast, effective data attribution method for neural networks in PyTorch

MadryLab/robustness

A library for experimenting with, training and evaluating neural networks, with a focus on adversarial robustness.

MadryLab/mnist_challenge

A challenge to explore adversarial robustness of neural networks on MNIST.

MadryLab/photoguard

Raising the Cost of Malicious AI-Powered Image Editing

MadryLab/trak

A fast, effective data attribution method for neural networks in PyTorch

MadryLab/implementation-matters

MadryLab/EditingClassifiers

MadryLab/robust-features-code

Code for "Robustness May Be at Odds with Accuracy"

MadryLab/datamodels-data

Data for "Datamodels: Predicting Predictions with Training Data"

MadryLab/modeldiff

ModelDiff: A Framework for Comparing Learning Algorithms

MadryLab/cox

A lightweight experimental logging library

MadryLab/modelcomponents

Decomposing and Editing Predictions by Modeling Model Computation

MadryLab/dataset-interfaces

Dataset Interfaces: Diagnosing Model Failures Using Controllable Counterfactual Generation

MadryLab/failure-directions

Distilling Model Failures as Directions in Latent Space

MadryLab/context-cite

Attribute (or cite) statements generated by LLMs back to in-context information.

MadryLab/DsDm

MadryLab/data-transfer

MadryLab/relu_stable

MadryLab/datamodels

MadryLab/journey-TRAK

Code for the paper "The Journey, Not the Destination: How Data Guides Diffusion Models"

MadryLab/rethinking-backdoor-attacks

MadryLab/bias-transfer

MadryLab/pretraining-distribution-shift-robustness

MadryLab/fast_l1

MadryLab/rla

Residue Level Alignment

MadryLab/AIaaS_Supply_Chains

Dataset and overview

MadryLab/post--adv-discussion

MadryLab/D3M

Debiasing Through Data Attribution

MadryLab/mirrorSDF

MadryLab/places-finetuning

MadryLab/trak_transfer_internal