Mediaspace scheduled maintenance: Aug 25, 2026 07:00 - 12:00 AM. During this time, videos will be temporarily unavailable. Check status updates.

Concept

Période iconoclaste de l'Empire byzantin

Publications associées (13)

Cross-chain payment protocols with success guarantees

Vincent Gramoli

In this paper, we consider the problem of cross-chain payment whereby customers of different escrows-implemented by a bank or a blockchain smart contract-successfully transfer digital assets without trusting each other. Prior to this work, cross-chain paym ...

SPRINGER2023

Democratizing Machine Learning

Nirupam Gupta, Rafaël Benjamin Pinot, Alexandre David Olivier Maurer

The increasing prevalence of personal devices motivates the design of algorithms that can leverage their computing power, together with the data they generate, in order to build privacy-preserving and effective machine learning models. However, traditional ...

IEEE2022

Genuinely distributed Byzantine machine learning

Rachid Guerraoui, El Mahdi El Mhamdi, Sébastien Louis Alexandre Rouault, Le Nguyen Hoang, Arsany Hany Abdelmessih Guirguis

Machine learning (ML) solutions are nowadays distributed, according to the so-called server/worker architecture. One server holds the model parameters while several workers train the model. Clearly, such architecture is prone to various types of component ...

2022

Leaderless Consensus

Rachid Guerraoui, Vincent Gramoli, Mihail Igor Zablotchi, Karolos Antoniadis

Classical synchronous consensus algorithms are leaderless: processes exchange their proposals, retain the maximum value and decide when they see the same choice across a couple of rounds. Indulgent consensus algorithms are more robust in that they only req ...

IEEE COMPUTER SOC2021

AKSEL: Fast Byzantine SGD

Rachid Guerraoui, El Mahdi El Mhamdi, Alexandre David Olivier Maurer, Sébastien Louis Alexandre Rouault

Modern machine learning architectures distinguish servers and workers. Typically, a d-dimensional model is hosted by a server and trained by n workers, using a distributed stochastic gradient descent (SGD) optimization scheme. At each SGD step, the goal is ...

Schloss Dagstuhl--Leibniz-Zentrum für Informatik2021

Genuinely Distributed Byzantine Machine Learning

Rachid Guerraoui, El Mahdi El Mhamdi, Sébastien Louis Alexandre Rouault, Le Nguyen Hoang, Arsany Hany Abdelmessih Guirguis

Machine Learning (ML) solutions are nowadays distributed, according to the so-called server/worker architecture. One server holds the model parameters while several workers train the model. Clearly, such architecture is prone to various types of component ...

Association for Computing Machinery2020

The Performance of Byzantine Fault Tolerant Blockchains

Vincent Gramoli

Blockchains have captured the attention of many, resulting in an abundance of new systems available for use. However, selecting an appropriate blockchain for an application is challenging due to the lack of comparative information discussing core metrics s ...

IEEE2020

AGGREGATHOR: Byzantine Machine Learning via Robust Gradient Aggregation

Rachid Guerraoui, El Mahdi El Mhamdi, Sébastien Louis Alexandre Rouault, Arsany Hany Abdelmessih Guirguis, Georgios Damaskinos

We present AGGREGATHOR, a framework that implements state-of-the-art robust (Byzantine-resilient) distributed stochastic gradient descent. Following the standard parameter server model, we assume that a minority of worker machines can be controlled by an a ...

2019

Scalable Byzantine Reliable Broadcast

Rachid Guerraoui, Matej Pavlovic, Dragos-Adrian Seredinschi, Matteo Monti

Byzantine reliable broadcast is a powerful primitive that allows a set of processes to agree on a message from a designated sender, even if some processes (including the sender) are Byzantine. Existing broadcast protocols for this setting scale poorly, as ...

Schloss Dagstuhl--Leibniz-Zentrum fer Informatik2019

Asynchronous Byzantine Machine Learning (the case of SGD)

Rachid Guerraoui, El Mahdi El Mhamdi, Mahsa Taziki, Rhicheek Patra, Georgios Damaskinos

Asynchronous distributed machine learning solutions have proven very effective so far, but always assuming perfectly functioning workers. In practice, some of the workers can however exhibit Byzantine behavior, caused by hardware failures, software bugs, c ...

2018

Names Trump Malice: Tiny Mobile Agents Can Tolerate Byzantine Failures

Rachid Guerraoui

We introduce a new theoretical model of ad hoc mobile computing in which agents have severely restricted memory, highly unpredictable movement and no initial knowledge of the system. Each agent’s memory can store O(1) bits, plus a unique identiﬁer, and O(1 ...

2009

On Locating Byzantine Attackers

Christina Fragouli, Suhas Diggavi, Mahdi Jafari Siavoshani

We examine networks that employ network coding, and are subject to Byzantine attacks. We consider systems where an appropriate network error correcting scheme is employed that is able to correct (up to a certain number of) Byzantine errors. Given this setu ...

2008

The Perfectly-Synchronized Round-based Model of Distributed Computing

Rachid Guerraoui, Bastian Pochon

The perfectly-synchronized round-based model provides the powerful abstraction of crash-stop failures with atomic and synchronous message delivery. This abstraction makes distributed programming very easy.We describe a technique to automatically transform ...

2007