Concept

Intoxication

Publications associées (10)

Machine Learning Security Against Data Poisoning: Are We There Yet?

Poisoning attacks compromise the training data utilized to train machine learning (ML) models, diminishing their overall performance, manipulating predictions on specific test samples, and implanting backdoors. This article thoughtfully explores these atta ...

Ieee Computer Soc2024

Robust Collaborative Learning with Linear Gradient Overhead

Rachid Guerraoui, Nirupam Gupta, John Stephan, Sadegh Farhadkhani, Le Nguyen Hoang, Rafaël Benjamin Pinot

Collaborative learning algorithms, such as distributed SGD (or D-SGD), are prone to faulty machines that may deviate from their prescribed algorithm because of software or hardware bugs, poisoned data or malicious behaviors. While many solutions have been ...

PLMR2023

An Equivalence Between Data Poisoning and Byzantine Gradient Attacks

Rachid Guerraoui, Sadegh Farhadkhani, Oscar Jean Olivier Villemaud, Le Nguyen Hoang

To study the resilience of distributed learning, the “Byzantine" literature considers a strong threat model where workers can report arbitrary gradients to the parameter server. Whereas this model helped obtain several fundamental results, it has sometimes ...

PMLR2022

On the relevance of P poisoning in real-world DOC aging

Davide Ferri

Understanding the real aging of Diesel oxidation catalysts (DOC) is key to predict their useful lifetime. This work comprises a detailed study of a commercial DOC in its unused state and after use in vehicle for 288,000 km. The activity of the samples was ...

ELSEVIER2021

MRI-based brain volumetry and retinal optical coherence tomography as the biomarkers of outcome in acute methanol poisoning

Tobias Kober, Bénédicte Marie Maréchal

Background: Basal ganglia lesions are typical findings on magnetic resonance imaging (MRI) of the brain in survivors of acute methanol poisoning. However, no data are available on the association between the magnitude of damaged brain regions, serum concen ...

ELSEVIER2020

Robust Distributed Learning

El Mahdi El Mhamdi

Whether it occurs in artificial or biological substrates, {\it learning} is a {distributed} phenomenon in at least two aspects. First, meaningful data and experiences are rarely found in one location, hence {\it learners} have a strong incentive to work t ...

EPFL2020

Methanol Poisoning as an Acute Toxicological Basal Ganglia Lesion Model: Evidence from Brain Volumetry and Cognition

Tobias Kober, Bénédicte Marie Maréchal

Background Acute methanol poisoning leads to optic neuropathy and necrotic lesions of basal ganglia (BG) and subcortical white matter. Survivors of methanol poisoning exhibit long-term executive and memory deficits. Associations between brain volumetry par ...

WILEY2019

Damage of Siloxanes on Ni-YSZ Anode Supported SOFC Operated on Hydrogen and Bio-Syngas

Jan Van Herle, Christian Ludwig, Stefan Diethelm, Hossein Madi

This work presents the poisoning effect of organic oxo-silicon compounds (siloxanes) which are generally found in sewage biogas. Lifetime and durability associated with solid oxide fuel cell (SOFC) technology is strongly related to the amount of contaminan ...

Wiley-V C H Verlag Gmbh2015

Combined Cr and S poisoning in solid oxide fuel cell cathodes

Jan Van Herle, Aïcha Hessler-Wyser, Quentin Thomas Jeangros, Zacharie Wuillemin, Josef Andreas Schuler, Caroline Calderone

This work aims to compare the effect of combined chromium and sulfur contaminating conditions to the Cr contamination alone on the Cr poisoning mechanisms in (La,Sr)MnO3-(Zr,Y)O-2 solid oxide fuel cell (SOFC) cathodes. Whereas Cr2O3 and (Cr,Mn)(3)O-4 are f ...

Elsevier2012

Cathode thickness-dependent tolerance to Cr-poisoning in solid oxide fuel cells

Christos Comninellis, Jan Van Herle, Josef Andreas Schuler, Pietro Tanasini

This work aimed to set new guidelines for the quantification of Cr accumulation in solid oxide fuel cell cathodes after operation, and enabled to pinpoint a diffusion-controlled tolerance to Cr-poisoning for increased cathode thickness; the additional cath ...

Elsevier2010

Graph Chatbot

Chattez avec Graph Search

Posez n’importe quelle question sur les cours, conférences, exercices, recherches, actualités, etc. de l’EPFL ou essayez les exemples de questions ci-dessous.

AVERTISSEMENT : Le chatbot Graph n'est pas programmé pour fournir des réponses explicites ou catégoriques à vos questions. Il transforme plutôt vos questions en demandes API qui sont distribuées aux différents services informatiques officiellement administrés par l'EPFL. Son but est uniquement de collecter et de recommander des références pertinentes à des contenus que vous pouvez explorer pour vous aider à répondre à vos questions.

Connectez-vous pour utiliser Chat avec Graph Search