MetaPoison: Practical General-purpose Clean-label Data Poisoning

Wei Huang, Jonas Geiping, Liam Fowl, +2

arXiv (Cornell University)

Artificial IntelligenceBilevel OptimizationEngineeringMachine LearningMeta-learning +16

Adversarially Robust Few-Shot Learning: A Meta-Learning Approach

Micah Goldblum, Liam Fowl, Tom Goldstein

arXiv (Cornell University)

Artificial IntelligenceFew-shot LearningImage ClassificationRobust Neural NetworksEngineering +13

Witches' Brew: Industrial Scale Data Poisoning via Gradient Matching

Jonas Geiping, Liam Fowl, Wei Huang, +4

arXiv (Cornell University)

Artificial IntelligenceData Poisoning AttacksConvolutional Neural NetworkEngineeringMachine Learning +17

Adversarial Examples Make Strong Poisons

Liam Fowl, Micah Goldblum, Ping-yeh Chiang, +3

arXiv (Cornell University)

Artificial IntelligenceAdversarial ExamplesAdversarial PoisoningMachine LearningData Science +13

Understanding Generalization through Visualizations

Wei Huang, Zeyad Emam, Micah Goldblum, +4

arXiv (Cornell University)

Artificial IntelligenceGeometric LearningEngineeringMachine LearningVisualization (Graphics) +14

Unraveling Meta-Learning: Understanding Feature Representations for Few-Shot Tasks

Micah Goldblum, Steven Reich, Liam Fowl, +3

arXiv (Cornell University)

Artificial IntelligenceFew-shot LearningEngineeringMachine LearningMeta-learning +14

Deep k-NN Defense Against Clean-Label Data Poisoning Attacks

Neehar Peri, Neal Gupta, Wei Huang, +5

Lecture notes in computer science

DP-InstaHide: Provably Defusing Poisoning and Backdoor Attacks with Differentially Private Data Augmentations

Eitan Borgnia, Jonas Geiping, Valeriia Cherepanova, +6

arXiv (Cornell University)

Privacy ProtectionEngineeringMachine LearningInformation SecurityInformation Forensics +17

Fishing for User Data in Large-Batch Federated Learning via Gradient Magnification

Yuxin Wen, Jonas Geiping, Liam Fowl, +2

arXiv (Cornell University)

Artificial IntelligenceEngineeringMachine LearningInformation SecurityFederated Structure +20

What Doesn't Kill You Makes You Robust(er): Adversarial Training against Poisons and Backdoors.

Jonas Geiping, Liam Fowl, Gowthami Somepalli, +3

arXiv (Cornell University)

Artificial IntelligenceEngineeringMachine LearningEvasion TechniqueInformation Security +16