Publications associées à Test de Wald

Machine Learning Security in Industry: A Quantitative Survey

Despite the large body of academic work on machine learning security, little is known about the occurrence of attacks on machine learning systems in the wild. In this paper, we report on a quantitative study with 139 industrial practitioners. We analyze at ...

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC2023

Incremental Learning in Diagonal Linear Networks

Raphaël Jean Berthier

Diagonal linear networks (DLNs) are a toy simplification of artificial neural networks; they consist in a quadratic reparametrization of linear regression inducing a sparse implicit regularization. In this paper, we describe the trajectory of the gradient ...

Microtome Publishing2023

A note on universal inference

Anthony Christopher Davison, Timmy Rong Tian Tse

Universal inference enables the construction of confidence intervals and tests without regularity conditions by splitting the data into two parts and appealing to Markov's inequality. Previous investigations have shown that the cost of this generality is a ...

WILEY2022

Experimental characterisation of textile compaction response: A benchmark exercise

Véronique Michaud, Baris Çaglar, Christophe Lopez

This paper reports the results of an international benchmark exercise on the measurement of fibre bed compaction behaviour. The aim was to identify aspects of the test method critical to obtain reliable results and to arrive at a recommended test procedure ...

2021

NovelT2Mapping for Evaluating Cervical Cancer Features by Providing QuantitativeT2Maps and Synthetic Morphologic Images: A Preliminary Study

Tobias Kober, Tom Hilbert, Yong Zhang, Meng Yang, Jie Liu

Background The application value of T(2)mapping in evaluating cervical cancer (CC) features remains unclear. Purpose To investigate the role of T(2)values in evaluating CC classification, grade, and lymphovascular space invasion (LVSI) in comparison to app ...

WILEY2020

On long time behavior of solutions to nonlinear dispersive equations

Stefano Francesco Burzio

In part I, we address the issue of existence of solutions for Cauchy problems involving nonlinear hyperbolic equations for initial data in Sobolev spaces with scaling subcritical regularity. In particular, we analyse nonlinear estimates for null-forms in t ...

EPFL2020

Learning new physics from a machine

Raffaele Tito D'Agnolo, Andrea Wulzer

We propose using neural networks to detect data departures from a given reference model, with no prior bias on the nature of the new physics responsible for the discrepancy. The virtues of neural networks as unbiased function approximants make them particu ...

American Physical Society2019

The detection of two-component mixture alternatives : Theory and methods

Daria Rukina

In this thesis, we deal with one of the facets of the statistical detection problem. We study a particular type of alternative, the mixture model. We consider testing where the null hypothesis corresponds to the absence of a signal, represented by some kno ...

EPFL2018

MAGIC observations of the February 2014 flare of 1ES 1011+496 and ensuing constraint of the EBL density

Thomas Günther Bretz

Context. During February-March 2014, the MAGIC telescopes observed the high-frequency peaked BL Lac 1ES 1011 + 496 (z = 0.212) in flaring state at very-high energy (VHE, E > 100 GeV). The flux reached a level of more than ten times higher than any previous ...

Edp Sciences S A2016

Comparing connectomes across subjects and populations at different scales

Jean-Philippe Thiran, Stephan Morgenthaler, Alessandra Griffa, Elda Fischi Gomez, Patric Hagmann, Djalel Eddine Meskaldji

Brain connectivity can be represented by a network that enables the comparison of the different patterns of struc- tural and functional connectivity among individuals. In the literature, two levels ofstatistical analysis have been con- sidered in comparing ...

2013

Determination of the X(3872) Meson Quantum Numbers

Olivier Schneider, Aurelio Bay, Yiming Li, Guido Haefeli, Frédéric Blanc, Christoph Frei, Tatsuya Nakada, Julien Rouvinet, Barinjaka Rakotomiaramanana, Raluca Anca Muresan, Michel De Cian, Christophe Salzmann, Johan Luisier, Anne Keune, Jian Wang, Vincent Fave, Yi Zhang, Lei Zhang, Jessica Prisciandaro, Mark Tobin, Minh Tâm Tran, Niko Neufeld, Matthew Needham, Marc-Olivier Bettler, Greig Alan Cowan, Maurizio Martinelli, Vladislav Balagura, Donal Patrick Hill, Cédric Potterat, Jean Wicht, Ilya Komarov, Plamen Hristov Hopchev, Bastien Luca Muster, Frédéric Guillaume Dupertuis, Pierre Jaton, Joël Bressieux, Songmei Wu, Albert Puig Navarro, Hans Dijkstra, Gerhard Raven, Peter Clarke, Frédéric Teubert, , , ,

Bivariate Extreme Statistics, Ii

We review the current state of statistical modeling of asymptotically independent data. Our discussion includes necessary and sufficient conditions for asymptotic independence, results on the asymptotic independence of statistics of interest, estimation an ...

2012

Modelling human perception of static facial expressions

Michel Bierlaire, Jean-Philippe Thiran, Gianluca Antonini, Javier Cruz Mota, Thomas Robin, Matteo Sorci

A recent internet based survey of over 35'000 samples has shown that when different human observers are asked to assign labels to static human facial expressions, different individuals categorize differently the same image. This fact results in a lack of a ...

2010

Impulse Radio Ultra-Wideband Ranging Algorithm under Multi User Environments

Jean-Yves Le Boudec, John Farserotu, Hai Zhan

Practical impulse radio ultra-wideband (IR-UWB) ranging systems always have to work in multi-user and weak non-line-of-sight (NLOS) environments. In this paper, we derive a novel IR-UWB ranging estimator under multi-user and weak NLOS environments. We mode ...

IEEE2009

Impulse Radio Ultra-WideBand ranging and localization for sensor networks

Hai Zhan

In this thesis, we focus on Impulse Radio (IR) Ultra-WideBand (UWB) ranging and positioning techniques under indoor propagation environments. IR-UWB, a new carrierless communication scheme using impulses, is a candidate technology for future communication, ...

EPFL2009

Impulse Radio Ultra-Wideband Ranging Based on Maximum Likelihood Estimation

Jean-Yves Le Boudec, John Farserotu, Hai Zhan

We propose a high-resolution ranging algorithm for impulse radio (IR) ultra-WideBand (UWB) communication systems in additive white Gaussian noise. We formulate the ranging problem as a maximum- likelihood (ML) estimation problem for the channel delays and ...

Institute of Electrical and Electronics Engineers2009

High-Resolution Impulse Radio Ultra Wideband Ranging

Jean-Yves Le Boudec, John Farserotu, Hai Zhan

We propose a high resolution ranging algorithm for unsynchronized impulse radio Ultra-wideband (UWB) communication systems in gaussian noise. We pose the ranging problem as a Maximum Likelihood (ML) estimation problem for the channel delays and attenuation ...

2007

A Novel Maximum Likelihood Estimation Of Superimposed Exponential Signals In Noise And Ultra-Wideband Application

Jean-Yves Le Boudec, John Farserotu, Hai Zhan

We pose the estimation of the parameters of multiple superimposed exponential signals in additive Gaussian noise problem as a Maximum Likelihood (ML) estimation problem. The ML problem is very non linear and hard to solve. Some previous works focused on fi ...

2007

The ecology of testate amoebae (Protists) in South-Central Alaska peatlands: building transfer function models for paleoenvironmental studies

Edward Mitchell

Testate amoebae are valuable indicators of peatland hydrology and have been used in many palaeoclimatic studies in peatlands. Because the species' ecological optima may vary around the globe, the development of transfer function models is an essential prer ...

2006

Optimal Conditionally Unbiased Bounded-Influence Inference in Dynamic Location and Scale Models

Loriano Mancini, Fabio Trojani

This paper studies the local robustness of estimators and tests for the conditional location and scale parameters in a strictly stationary time series model. We first derive optimal bounded-influence estimators for such settings under a conditionally Gauss ...

2005