Picture for El-Mahdi El-Mhamdi

El-Mahdi El-Mhamdi

Inverting Gradient Attacks Naturally Makes Data Poisons: An Availability Attack on Neural Networks

Add code
Oct 28, 2024
Viaarxiv icon

On Goodhart's law, with an application to value alignment

Add code
Oct 12, 2024
Viaarxiv icon

Data Taggants: Dataset Ownership Verification via Harmless Targeted Data Poisoning

Add code
Oct 09, 2024
Viaarxiv icon

SoK: On the Impossible Security of Very Large Foundation Models

Add code
Sep 30, 2022
Viaarxiv icon

Garfield: System Support for Byzantine Machine Learning

Add code
Oct 12, 2020
Figure 1 for Garfield: System Support for Byzantine Machine Learning
Figure 2 for Garfield: System Support for Byzantine Machine Learning
Figure 3 for Garfield: System Support for Byzantine Machine Learning
Figure 4 for Garfield: System Support for Byzantine Machine Learning
Viaarxiv icon

Collaborative Learning as an Agreement Problem

Add code
Aug 04, 2020
Figure 1 for Collaborative Learning as an Agreement Problem
Viaarxiv icon

Distributed Momentum for Byzantine-resilient Learning

Add code
Mar 09, 2020
Figure 1 for Distributed Momentum for Byzantine-resilient Learning
Figure 2 for Distributed Momentum for Byzantine-resilient Learning
Figure 3 for Distributed Momentum for Byzantine-resilient Learning
Figure 4 for Distributed Momentum for Byzantine-resilient Learning
Viaarxiv icon

Fast and Secure Distributed Learning in High Dimension

Add code
May 05, 2019
Figure 1 for Fast and Secure Distributed Learning in High Dimension
Figure 2 for Fast and Secure Distributed Learning in High Dimension
Figure 3 for Fast and Secure Distributed Learning in High Dimension
Viaarxiv icon

SGD: Decentralized Byzantine Resilience

Add code
May 05, 2019
Figure 1 for SGD: Decentralized Byzantine Resilience
Figure 2 for SGD: Decentralized Byzantine Resilience
Figure 3 for SGD: Decentralized Byzantine Resilience
Figure 4 for SGD: Decentralized Byzantine Resilience
Viaarxiv icon