Publications associées à GARFIELD: System Support for Byzantine Machine Learning (Regular Paper)

Planetary-Scale Byzantine Fault Tolerance

The scale and pervasiveness of the Internet make it a pillar of planetary communication, industry and economy, as well as a fundamental medium for public discourse and democratic engagement. In stark contrast with the Internet's decentralized infrastructur ...

EPFL2024

Live-cell imaging powered by computation

Suliana Manley, Florian Jug

The proliferation of microscopy methods for live-cell imaging offers many new possibilities for users but can also be challenging to navigate. The prevailing challenge in live-cell fluorescence microscopy is capturing intra-cellular dynamics while preservi ...

Nature Portfolio2024

Intermediate Address Space: virtual memory optimization of heterogeneous architectures for cache-resident workloads

David Atienza Alonso, Marina Zapater Sancho, Luis Maria Costero Valero, Darong Huang, Qunyou Liu

The increasing demand for computing power and the emergence of heterogeneous computing architectures have driven the exploration of innovative techniques to address current limitations in both the compute and memory subsystems. One such solution is the use ...

2024

Fixing by Mixing: A Recipe for Optimal Byzantine ML under Heterogeneity

Rachid Guerraoui, John Stephan, Sadegh Farhadkhani, Nirupam Gupta, Rafaël Benjamin Pinot, Youssef Allouah

Byzantine machine learning (ML) aims to ensure the resilience of distributed learning algorithms to misbehaving (or Byzantine) machines. Although this problem received significant attention, prior works often assume the data held by the machines to be homo ...

PMLR2023

Privatized graph federated learning

Ali H. Sayed, Stefan Vlaski, Elsa Rizk

Federated learning is a semi-distributed algorithm, where a server communicates with multiple dispersed clients to learn a global model. The federated architecture is not robust and is sensitive to communication and computational overloads due to its one-m ...

SPRINGER2023

Impact of Redundancy on Resilience in Distributed Optimization and Learning

Shuo Liu, Nirupam Gupta

This paper considers the problem of resilient distributed optimization and stochastic learning in a server-based architecture. The system comprises a server and multiple agents, where each agent has its own local cost function. The agents collaborate with ...

Assoc Computing Machinery2023

Rebooting Virtual Memory with Midgard

Siddharth Gupta

Virtual Memory (VM) is a critical programming abstraction that is widely used in various modern computing platforms. With the rise of datacenter computing and birth of planet-scale online services, the semantic and capacity requirements from memory have ev ...

EPFL2023

Scalable Computation of Dynamic Flow Problems via Multimarginal Graph-Structured Optimal Transport

Axel Ringh, Yiqiang Chen, Isabel Haasler

In this work, we develop a new framework for dynamic network flow pro-blems based on optimal transport theory. We show that the dynamic multicommodity minimum-cost network flow problem can be formulated as a multimarginal optimal transport problem, where t ...

2023

Byzantine Machine Learning Made Easy By Resilient Averaging of Momentums

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

Byzantine resilience emerged as a prominent topic within the distributed machine learning community. Essentially, the goal is to enhance distributed optimization algorithms, such as distributed SGD, in a way that guarantees convergence despite the presence ...

PMLR2022

Genuinely distributed Byzantine machine learning

Rachid Guerraoui, El Mahdi El Mhamdi, Le Nguyen Hoang, Sébastien Louis Alexandre Rouault, 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

Hardware and Software Support for RPC-Centric Server Architecture

Mark Johnathon Sutherland

Online services have become ubiquitous in technological society, the global demand for which has driven enterprises to construct gigantic datacenters that run their software. Such facilities have also recently become a substrate for third-party organizatio ...

EPFL2022

Oracular Byzantine Reliable Broadcast

Rachid Guerraoui, Matteo Monti, Manuel José Ribeiro Vidigueira, Martina Camaioni

Byzantine Reliable Broadcast (BRB) is a fundamental distributed computing primitive, with applications ranging from notifications to asynchronous payment systems. Motivated by practical consideration, we study Client-Server Byzantine Reliable Broadcast (CS ...

Schloss Dagstuhl - Leibniz-Zentrum für Informatik2022

An Equivalence Between Data Poisoning and Byzantine Gradient Attacks

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

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 advantages of P2P ML on mobile devices

Rachid Guerraoui, Alexandre David Olivier Maurer

Many fields make use nowadays of machine learning (ML) enhanced applications for cost optimization, scheduling or forecasting, in- cluding the energy sector. However, these very ML algorithms consume a significant amount of energy, sometimes going against ...

ACM2022

Quantization for Distributed Processing and Learning of Structured Data

Isabela Cunha Maia Nobre

In the domains of machine learning, data science and signal processing, graph or network data, is becoming increasingly popular. It represents a large portion of the data in computer, transportation systems, energy networks, social, biological, and other s ...

EPFL2022

When to Hedge in Interactive Services

Aikaterini Argyraki, Edouard Bugnion, Mia Primorac

In online data-intensive (OLDI) services, each client request typically executes on multiple servers in parallel; as a result, “system hiccups”, although rare within a single server, can interfere with many client requests and cause violations of service-l ...

USENIX ASSOC2021

Collaborative Learning in the Jungle (Decentralized, Byzantine, Heterogeneous, Asynchronous and Nonconvex Learning)

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

We study Byzantine collaborative learning, where n nodes seek to collectively learn from each others' local data. The data distribution may vary from one node to another. No node is trusted, and f < n nodes can behave arbitrarily. We prove that collaborati ...

2021

A Coarse-mesh Methodology for the Analysis of One and Two-phase Nuclear Reactor Thermal-hydraulics in a Multi-physics Context

Stefan Radman

The analysis of nuclear reactors for performance and safety assessment benefits from the use of computational tools. In this context, this work aims at the development and application of a thermal-hydraulics methodology and related software that respond to ...

EPFL2021

Rational Agreement in the Presence of Crash Faults

Vincent Gramoli, Alejandro Ranchal Pedrosa

Blockchain systems need to solve consensus despite the presence of rational users and failures. The notion of (k, t)-robustness is key to derive impossibility results with k rational players and t faulty players. However, these t faulty players are always ...

IEEE2021

Dynamic Byzantine Reliable Broadcast

Rachid Guerraoui, Dragos-Adrian Seredinschi, Jovan Komatovic, Andrei Tonkikh

Reliable broadcast is a communication primitive guaranteeing, intuitively, that all processes in a distributed system deliver the same set of messages. The reason why this primitive is appealing is twofold: (i) we can implement it deterministically in a co ...

Schloss Dagstuhl--Leibniz-Zentrum f{\"u}r Informatik2021