Publications associées à Architecture byzantine

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

Learning from Chernobyl: Reverse Colonialism and Feral Architectures

Aitor Frías Sanchez, Olena Orap, Joaquin Perailes Santiago

This article revolves around the concept of reverse colonialism and uses the fascinating example of Pripyat as a case study. Pripyat was deeply affected by the devastating Chornobyl nuclear accident in 1986. Following the disaster, the disappearance o ...

2023

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

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

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

Reuse of Load-Bearing Timber Elements - Case Study of a Lookout Tower in Lausanne, Switzerland

Corentin Jean Dominique Fivet, Jonas Warmuth, Edisson Xavier Estrella Arcos, Julius Natterer, Benoît Pascal Kälin

The construction industry is jeopardizing Earth’s environment; current mitigation efforts do not suffice; and additional sustainable strategies are required. One such strategy consists of reusing structural building elements over multiple service lives. In ...

Curran Associates, Inc.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

An Equivalence Between Data Poisoning and Byzantine Gradient Attacks

Rachid Guerraoui, Sadegh Farhadkhani, Le Nguyen Hoang, 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

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

Rachid Guerraoui, Sébastien Louis Alexandre Rouault, Arsany Hany Abdelmessih Guirguis

We present GARFIELD, a library to transparently make machine learning (ML) applications, initially built with popular (but fragile) frameworks, e.g., TensorFlow and PyTorch, Byzantine–resilient. GARFIELD relies on a novel object–oriented design, reducing t ...

IEEE2021

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

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

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

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

The New Brutalism, invenzione di uno stile

Silvia Groaz

The concept of New Brutalism has been the victim of a series of misunderstandings, myths and aporias. The initial intentions, devoid of any principle or dogma, have passed from hand to hand. The original definitionâ if only one can be identifiedâ has be ...

EPFL2021

Dynamic Byzantine Reliable Broadcast

Rachid Guerraoui, Jovan Komatovic, Dragos-Adrian Seredinschi, 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

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 ...

Association for Computing Machinery2020

Scalable Byzantine Reliable Broadcast

Rachid Guerraoui, Dragos-Adrian Seredinschi, Matej Pavlovic, 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

Les "vestiges" de l'opération Million dans l'oeuvre de Georges Candilis - Actualités et stratégies de sauvegarde pour un patrimoine de l'habitat économique du second après-guerre

Korinna Zinovia Weber

The preservation of post-war architecture represents a complex challenge to overcome for different reasons: hesitating recognition, bad perception and a certain material fragility. Especially the economic post-war housing has difficulties to be recognized ...

EPFL2019

Plug-and-Play Fault Detection and Isolation for Large-Scale Nonlinear Systems With Stochastic Uncertainties

Giancarlo Ferrari Trecate

This paper proposes a novel scalable model-based fault detection and isolation approach for the monitoring of nonlinear large-scale systems, consisting of a network of interconnected subsystems. The fault diagnosis architecture is designed to automatically ...

2019

Building Duration / El Portico de los Huespedes

Patricia Guaita, Raffael Baur, Sony Devabhaktuni

El Portico de los Huéspedes is a pedagogical, design-research project that tests the possibility of an open-ended architecture, where duration links between past and future. It is a work of architecture as becoming: ‘work’ meaning the structure, rhythms an ...

Association of Collegiate Schools of Architecture2019

Scaling Byzantine Fault Tolerance

Matej Pavlovic

Online services are becoming more and more ubiquitous and keep growing in scale. At the same time, they are required to be highly available, secure, energy-efficient, and to achieve high performance. To ensure these (and many other) properties, replication ...

EPFL2019

The Hidden Vulnerability of Distributed Learning in Byzantium

Rachid Guerraoui, El Mahdi El Mhamdi, Sébastien Louis Alexandre Rouault

While machine learning is going through an era of celebrated success, concerns have been raised about the vulnerability of its backbone: stochastic gradient descent (SGD). Recent approaches have been proposed to ensure the robustness of distributed SGD aga ...

2018

RAIN: A Bio-Inspired Communication and Data Storage Infrastructure

Matteo Monti

We summarize the results and perspectives from a companion article, where we presented and evaluated an alternative architecture for data storage in distributed networks. We name the bio-inspired architecture RAIN, and it offers file storage service that, ...

Mit Press2017

A General Framework for Architecture Composability

Joseph Sifakis, Simon Bliudze, Eduard Baranov

Architectures depict design principles: paradigms that can be understood by all, allow thinking on a higher plane and avoiding low-level mistakes. They provide means for ensuring correctness by construction by enforcing global properties characterizing the ...

Springer Verlag2016