Publications associées à UQGAN: A Unified Model for Uncertainty Quantification of Deep Classifiers trained via Conditional GANs

Publications associées (25)

Function-Space Optimality of Neural Architectures with Multivariate Nonlinearities

We investigate the function-space optimality (specifically, the Banach-space optimality) of a large class of shallow neural architectures with multivariate nonlinearities/activation functions. To that end, we construct a new family of Banach spaces defined ...

SIAM PUBLICATIONS2025

Surface distributions and vertical profiles of trace gases (CO, O3, NO, NO2) in the Arctic wintertime boundary layer using low-cost sensors during ALPACA-2022

Julia Schmale, Andrea Baccarini, Roman Pohorsky

Electrochemical gas sensors (EGSs) have been used to measure the surface distributions and vertical profiles of trace gases in the wintertime Arctic boundary layer during the Alaskan Layered Pollution and Chemical Analysis (ALPACA) field experiment in Fair ...

COPERNICUS GESELLSCHAFT MBH2025

Combinatorial Aspects of Continuous Optimization: Graph Polynomials and Neural Networks

Amire Bendjeddou

In this thesis, we study how combinatorial structures and symmetries emerge from analytic properties of graph polynomials and neural networks. Graph polynomials encode combinatorial aspects of the underlying graph in their coefficients and are remarkably u ...

EPFL2025

Ion Sieving in Two-Dimensional Membranes from First Principles

Nicola Marzari, Nicéphore Arthur François Bonnet

A first-principles approach for calculating ion separation in solution through two-dimensional (2D) membranes is proposed and applied. Ionic energy profiles across the membrane are obtained first, where solvation effects are simulated explicitly with machi ...

2025

Masked Diffusion Strategy for Privacy-Preserving Distributed Learning

Ali H. Sayed

To protect both local gradients and estimated parameters in distributed learning, this paper introduces a masked diffusion (MD) strategy, leading to two algorithms: the MD stochastic gradient (MD-SG) and the MD primal-dual stochastic gradient (MPD-SG). The ...

2025

Euclid: Searches for strong gravitational lenses using convolutional neural nets in Early Release Observations of the Perseus field

Georges Meylan, Benjamin Yvan Alexandre Clement, Olivia Mueller

The Euclid Wide Survey (EWS) is predicted to find approximately 170 000 galaxy-galaxy strong lenses from its lifetime observation of 14 000 deg2 of the sky. Detecting this many lenses by visual inspection with professional astronomers and citizen scientist ...

2025

Learning Hand State Estimation for a Light Exoskeleton

Luca Randazzo

We propose a machine learning-based estimator of the hand state for rehabilitation purposes, using light exoskeletons. These devices are easy to use and useful for delivering domestic and frequent therapies. We build a supervised approach using information ...

Springer Nature2025

How Two-Layer Neural Networks Learn, One (Giant) Step at a Time

Florent Gérard Krzakala, Bruno Loureiro, Ludovic Théo Stephan, Yatin Dandi, Luca Pesce

For high-dimensional Gaussian data, we investigate theoretically how the features of a two-layer neural network adapt to the structure of the target function through a few large batch gradient descent steps, leading to an improvement in the approximation c ...

2024

Euclid preparation XXXIII. Characterization of convolutional neural networks for the identification of galaxy-galaxy strong-lensing events

Frédéric Courbin, Georges Meylan, Georgios Vernardos, Benjamin Yvan Alexandre Clement

Forthcoming imaging surveys will increase the number of known galaxy-scale strong lenses by several orders of magnitude. For this to happen, images of billions of galaxies will have to be inspected to identify potential candidates. In this context, deep-le ...

Edp Sciences S A2024

Simultaneous Quantification of Multiple Drugs by Machine Learning on Electrochemical Sensors

Sandro Carrara, Lin Du, Tatsunori Matsumoto

Electrochemical devices and systems are significant for the development of therapeutic drug monitoring (TDM) and personalized therapy. However, electrochemical sensors are usually not that much selective. Therefore, innovative machine learning (ML) approac ...

Institute of Electrical and Electronics Engineers Inc.2024

A PAC-Bayesian Framework for Optimal Control with Stability Guarantees

Mahrokh Ghoddousiboroujeni, Giancarlo Ferrari Trecate, Clara Lucía Galimberti

Stochastic Nonlinear Optimal Control (SNOC) involves minimizing a cost function that averages out the random uncertainties affecting the dynamics of nonlinear systems. For tractability reasons, this problem is typically addressed by minimizing an empirical ...

IEEE2024

Rehabilitation exoskeleton system with bidirectional virtual reality feedback training strategy

Rui Wu

Virtual reality (VR) technology revitalises rehabilitation training by creating rich, interactive virtual rehabilitation scenes and tasks that deeply engage patients. Robotics with immersive VR environments have the potential to significantly enhance the s ...

2024

Fast refacing of MR images with a generative neural network lowers re-identification risk and preserves volumetric consistency

Jean-Philippe Thiran, Tobias Kober, Bénédicte Marie Maréchal

With the rise of open data, identifiability of individuals based on 3D renderings obtained from routine structural magnetic resonance imaging (MRI) scans of the head has become a growing privacy concern. To protect subject privacy, several algorithms have ...

Wiley2024

On-chip artificial neural network for PET source position reconstruction

Edoardo Charbon, Claudio Bruschini, Emanuele Ripiccini

This work presents the characterization results and methods used to determine the reconstruction of a 22Na point source placed in a coincidence setup. For this purpose, a novel, fully-integrated, reconfigurable feed-forward artificial neural network was im ...

IEEE2023

The dynamics of representation learning in shallow, non-linear autoencoders

Maria Refinetti

Autoencoders are the simplest neural network for unsupervised learning, and thus an ideal framework for studying feature learning. While a detailed understanding of the dynamics of linear autoencoders has recently been obtained, the study of non-linear aut ...

JMLR-JOURNAL MACHINE LEARNING RESEARCH2022

Advanced Techniques in Optical Diffraction Tomography

Ahmed Bassam Sayed Ayoub Mohamed Emam

In this thesis, we study the 3 challenges described above. First, we study different reconstruction techniques and assess the fidelity of each reconstruction results by means of structured illumination and phase conjugation. By reconstructing the 3D refrac ...

EPFL2022

Overcoming the Domain Gap in Neural Action Representations

Pascal Fua, Pavan P Ramdya, Semih Günel, Sina Honari, Florian Aymanns

Relating behavior to brain activity in animals is a fundamental goal in neuroscience, with practical applications in building robust brain-machine interfaces. However, the domain gap between individuals is a major issue that prevents the training of genera ...

SPRINGER2022

Physics-based representations for machine learning properties of chemical reactions

Anne-Clémence Corminboeuf, Matthew Wodrich, Alberto Fabrizio, Puck Elisabeth van Gerwen

Physics-based representations constructed using only atomic positions and nuclear charges (also known as quantum machine learning, QML) allow for the reliable and efficient inference of molecular properties from training data. Chemistry is a science rooted ...

2022

How isotropic kernels perform on simple invariants

Matthieu Wyart, Stefano Spigler, Jonas Paccolat

We investigate how the training curve of isotropic kernel methods depends on the symmetry of the task to be learned, in several settings. (i) We consider a regression task, where the target function is a Gaussian random field that depends only on d(paralle ...

IOP PUBLISHING LTD2021

Align, then memorise: the dynamics of learning with feedback alignment

Maria Refinetti

Direct Feedback Alignment (DFA) is emerging as an efficient and biologically plausible alternative to backpropagation for training deep neural networks. Despite relying on random feedback weights for the backward pass, DFA successfully trains state-of-the- ...

JMLR-JOURNAL MACHINE LEARNING RESEARCH2021