Publication
Fixing by Mixing: A Recipe for Optimal Byzantine ML under Heterogeneity
Publications associées (34)
Nanoparticles and method for producing uniform silicate-based nanoparticles
Nanoparticles and method for producing uniform silicate-based nanoparticles are disclosed. The method comprises a step of injecting into tubular branched elements comprising static mixers a first aqueous solution comprising a water-soluble silicate compoun ...
2018Mean-Field methods for Structured Deep-Learning in Computer Vision
In recent years, Machine Learning based Computer Vision techniques made impressive progress. These algorithms proved particularly efficient for image classification or detection of isolated objects. From a probabilistic perspective, these methods can predi ...
EPFL2018Asynchronous Byzantine Machine Learning (the case of SGD)
Rachid Guerraoui, Mahsa Taziki, El Mahdi El Mhamdi, 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 ...
2018Contained density currents with high volume of release
Mário Jorge Rodrigues Pereira Da Franca, Quentin Theiler
Contained density currents with high volume of release reflect against the boundaries of the reception environment commonly leading to oscillatory flow. These flows exist in sediment clarifiers, compromising their operations, and deposited signatures of co ...
Wiley-Blackwell2016Temporally Coherent Clustering of Student Data
The extraction of student behavior is an important task in educational data mining. A common approach to detect similar behavior patterns is to cluster sequential data. Standard approaches identify clusters at each time step separately and typically show l ...
EDM2016Beyond One Third Byzantine Failures
Rachid Guerraoui, Anne-Marie Kermarrec, Cheng Wang
The Byzantine agreement problem requires a set of processes to agree on a value sent by a transmitter, despite a subset of processes behaving in an arbitrary, i.e. Byzantine, manner and sending corrupted messages to all processes in the system. It ...
2015Cluster phenomenon and partial differential aging in RAP mixtures
André-Gilles Dumont, Michel Pittet, Sara Bressi
The present study defines a procedure for gaining a better understanding of the interaction between the aged and the virgin binders and what actually occurs during a mix process. Indeed, the authors propose a methodology to determine and quantify the clust ...
Elsevier Sci Ltd2015Semi-supervised and unsupervised kernel-based novelty detection with application to remote sensing images
Frank Grégoire Jean de Morsier
The main challenge of new information technologies is to retrieve intelligible information from the large volume of digital data gathered every day. Among the variety of existing data sources, the satellites continuously observing the surface of the Earth ...
EPFL2014Computational Methods For Structured Sparse Component Analysis of Convolutive Speech Mixtures
Hervé Bourlard, Volkan Cevher, Afsaneh Asaei
We cast the under-determined convolutive speech separation as sparse approximation of the spatial spectra of the mixing sources. In this framework we compare and contrast the major practical algorithms for structured sparse recovery of speech signal. Speci ...
2012Computational Methods For Structured Sparse Component Analysis of Convolutive Speech Mixtures
Hervé Bourlard, Volkan Cevher, Afsaneh Asaei
We cast the under-determined convolutive speech separation as sparse approximation of the spatial spectra of the mixing sources. In this framework we compare and contrast the major practical algorithms for structured sparse recovery of speech signal. Speci ...
2012