Publications associées à Loi sécante hyperbolique

Making Computer Vision Models Robust and Adaptive

Visual perception is indispensable for many real-world applications. However, perception models deployed in the real world will encounter numerous and unpredictable distribution shifts, for example, changes in geographic locations, motion blur, and adverse ...

EPFL2023

Learning Robust and Adaptive Representations: from Interactions, for Interactions

Yuejiang Liu

Interactions are ubiquitous in our world, spanning from social interactions between human individuals to physical interactions between robots and objects to mechanistic interactions among different components of an intelligent system. Despite their prevale ...

EPFL2023

Rapid Network Adaptation: Learning to Adapt Neural Networks Using Test-Time Feedback

Amir Roshan Zamir, Oguzhan Fatih Kar, Shuqing Teresa Yeo, Zahra Sodagar

We propose a method for adapting neural networks to distribution shifts at test-time. In contrast to training-time robustness mechanisms that attempt to anticipate and counter the shift, we create a closed-loop system and make use of test-time feedback sig ...

Ieee Computer Soc2023

Towards Robust Vision Transformer

Shaokai Ye, Yuan He

Recent advances on Vision Transformer (ViT) and its improved variants have shown that self-attention-based networks surpass traditional Convolutional Neural Networks (CNNs) in most vision tasks. However, existing ViTs focus on the standard accuracy and com ...

IEEE COMPUTER SOC2022

OptTTA: Learnable Test-Time Augmentation for Source-Free Medical Image Segmentation Under Domain Shift

Jean-Philippe Thiran, Guillaume Marc Georges Vray, Devavrat Tomar, Seyedbehzad Bozorgtabar

As distribution shifts are inescapable in realistic clinical scenarios due to inconsistencies in imaging protocols, scanner vendors, and across different centers, well-trained deep models incur a domain generalization problem in unseen environments. Despit ...

PMLR2022

Robustness and invariance properties of image classifiers

Apostolos Modas

Deep neural networks have achieved impressive results in many image classification tasks. However, since their performance is usually measured in controlled settings, it is important to ensure that their decisions remain correct when deployed in noisy envi ...

EPFL2022

Robust Generalization despite Distribution Shift via Minimum Discriminating Information

Daniel Kuhn, Tobias Sutter, Andreas Krause

Training models that perform well under distribution shifts is a central challenge in machine learning. In this paper, we introduce a modeling framework where, in addition to training data, we have partial structural knowledge of the shifted test distribut ...

2021

Redundant features can hurt robustness to distribution shift

Pascal Frossard, Seyed Mohsen Moosavi Dezfooli, Guillermo Ortiz Jimenez, Apostolos Modas

In this work, we borrow tools from the field of adversarial robustness, and propose a new framework that permits to relate dataset features to the distance of samples to the decision boundary. Using this framework we identify the subspace of features used ...

2020

Strong convergence of multivariate maxima

Simone Padoan, Stefano Rizzelli

It is well known and readily seen that the maximum of n independent and uniformly on [0, 1] distributed random variables, suitably standardised, converges in total variation distance, as n increases, to the standard negative exponential distribution. We ex ...

CAMBRIDGE UNIV PRESS2020

Impact of a companion and of chromospheric emission on the shape of chromosome maps for globular clusters

Carmela Lardo

Context. Globular clusters (GCs) host multiple populations of stars that are well-separated in a photometric diagram - the chromosome map - built from specific Hubble Space Telescope (HST) filters. Stars from different populations feature at various locati ...

EDP SCIENCES S A2020

Uncertainty and sensitivity analysis of PWR mini-core transients in the presence of nuclear data uncertainty using non-parametric tolerance limits

Andreas Pautz, Winfried Zwermann, Alexander Aures

The impact of nuclear data uncertainties is studied for the reactor power and the reactivity during control rod withdrawal transients with reactivity insertions of 0.5

and 0.97

, respectively, for a PWR mini-core model. Multi-group cross sections, the mul ...

PERGAMON-ELSEVIER SCIENCE LTD2020

35 years of photovoltaics: Analysis of the TISO-10-kW solar plant, lessons learnt in safety and performance-Part 1

Christophe Ballif, Alessandro Francesco Aldo Virtuani, Eleonora Annigoni

The TISO-10-kW solar plant, connected to the grid in 1982, is the oldest installation of this kind in Europe. Its history is well documented, and the full set of modules has been tested indoors at regular intervals over the years. After 35 years of operati ...

WILEY2019

Statistical inference in ensemble modeling of cellular metabolism

Vassily Hatzimanikatis, Marc-Olivier Boldi, Tuure Eelis Hameri

Kinetic models of metabolism can be constructed to predict cellular regulation and devise metabolic engineering strategies, and various promising computational workflows have been developed in recent years for this. Due to the uncertainty in the kinetic pa ...

2019

Galactic rotation curves versus ultralight dark matter: Implications of the soliton-host halo relation

Sergey Sibiryakov, Diego Blas Temino

Bosonic ultralight dark matter (ULDM) would form cored density distributions at the center of galaxies. These cores, seen in numerical simulations, admit analytic description as the lowest energy bound state solution ("soliton") of the Schroedinger-Poisson ...

American Physical Society2018

No Evidence for Association of beta-Defensin Genomic Copy Number with HIV Susceptibility, HIV Load during Clinical Latency, or Progression to AIDS

Jacques Fellay, Paul Jack Mc Laren

Common single-nucleotide variation in the host accounts for 25% of the variability in the plasma levels of HIV during the clinical latency stage (viral load set point). However, the role of rare variants and copy number variants remains relatively unexplor ...

Wiley-Blackwell2017

Robust Online Time Series Prediction with Recurrent Neural Networks

Karl Aberer, Tian Guo

Time series forecasting for streaming data plays an important role in many real applications, ranging from IoT systems, cyber-networks, to industrial systems and healthcare. However the real data is often complicated with anomalies and change points, which ...

Ieee2016

Probing the Limits of Social Data : Biases, Methods, and Domain Knowledge

Alexandra Olteanu

Online social data has been hailed to provide unprecedented insights into human phenomena due to its ability to capture human behavior at a scale and level of detail, both in breadth and depth, that is hard to achieve through conventional data collection t ...

EPFL2016

Multiplex networks in metropolitan areas: generic features and local effects

Emanuele Strano, Marc Barthélemy

Most large cities are spanned by more than one transportation system. These different modes of transport have usually been studied separately: it is however important to understand the impact on urban systems of coupling different modes and we report in th ...

Royal Soc2015

Simulation based population synthesis

Michel Bierlaire, Gunnar Flötteröd, Bilal Farooq, Ricardo Daniel Hurtubia González

Microsimulation of urban systems evolution requires synthetic population as a key input. Currently, the focus is on treating synthesis as a fitting problem and thus various techniques have been developed, including Iterative Proportional Fitting (IPF) and ...

Pergamon-Elsevier Science Ltd2013

Extended Uniform Distribution Accounting for Uncertainty of Uncertainty

Ian Smith, James-Alexandre Goulet

Increasingly, uncertainties are explicitly considered for important engineering tasks. Often, little case-specific information is available for characterizing these uncertainties. Uniform distributions are an easy way to describe errors in absence of more ...

ASCE2011