Publications associées à Connexionnisme

Byzantine-Resilient Learning Beyond Gradients: Distributing Evolutionary Search

Rachid Guerraoui, Andrei Kucharavy, Matteo Monti

Modern machine learning (ML) models are capable of impressive performances. However, their prowess is not due only to the improvements in their architecture and training algorithms but also to a drastic increase in computational power used to train them.|S ...

Assoc Computing Machinery2023

Kernel Memory Networks: A Unifying Framework for Memory Modeling

Wulfram Gerstner, Johanni Michael Brea, Georgios Iatropoulos

We consider the problem of training a neural network to store a set of patterns with maximal noise robustness. A solution, in terms of optimal weights and state update rules, is derived by training each individual neuron to perform either kernel classifica ...

Curran Associates2022

Strong temperature-dependent thermoelectric power of IrMn(3)thin films of different thicknesses

Jean-Philippe Ansermet, Haiming Yu, Junfeng Hu, Mukund Krishna Kothari

IrMn(3)thin films were experimentally characterized by thermoelectric power and resistivity as a function of temperature and sample thickness. The Seebeck coefficient shows strong temperature dependence and the value increases with temperature up to 350 K. ...

IOP PUBLISHING LTD2021

Node Attribute Completion in Knowledge Graphs with Multi-Relational Propagation

Robert West, Eda Bayram, Alberto Garcia Duran

The existing literature on knowledge graph completion mostly focuses on the link prediction task. However, knowledge graphs have an additional incompleteness problem: their nodes possess numerical attributes, whose values are often missing. Our approach, d ...

2021

Model Fusion via Optimal Transport

Martin Jaggi, Sidak Pal Singh

Combining different models is a widely used paradigm in machine learning applications. While the most common approach is to form an ensemble of models and average their individual predictions, this approach is often rendered infeasible by given resource co ...

2020

Architectural Sampling: A Formal Basis for Machine-Learnable Architecture

Immanuel Chee Beng Koh

The spatial and formal conception of architecture, and thus its modes of design perception and representation, directly contributes to its machine-learnability; and consequently, its capacity in leveraging today's machine learning apparatus for design inno ...

EPFL2019

Learning to Find Good Correspondences

Pascal Fua, Mathieu Salzmann, Eduard Trulls Fortuny, Vincent Lepetit, Kwang Moo Yi, Yuki Ono

We develop a deep architecture to learn to find good correspondences for wide-baseline stereo. Given a set of putative sparse matches and the camera intrinsics, we train our network in an end-to-end fashion to label the correspondences as inliers or outlie ...

2018

Learning to Find Good Correspondences

Pascal Fua, Mathieu Salzmann, Vincent Lepetit, Kwang Moo Yi, Eduard Trulls Fortuny, Yuki Ono

We develop a deep architecture to learn to find good correspondences for wide-baseline stereo. Given a set of putative sparse matches and the camera intrinsics, we train our network in an end-to-end fashion to label the correspondences as inliers or outlie ...

IEEE2018

Non-intrusive reduced order modeling of nonlinear problems using neural networks

Jan Sickmann Hesthaven, Stefano Ubbiali

We develop a non-intrusive reduced basis (RB) method for parametrized steady-state partial differential equations (PDEs). The method extracts a reduced basis from a collection of high-fidelity solutions via a proper orthogonal decomposition (POD) and emplo ...

2018

Unsupervised Learning of Sentence Embeddings using Compositional n-Gram Features

Martin Jaggi, Matteo Pagliardini, Prakhar Gupta

The recent tremendous success of unsupervised word embeddings in a multitude of applications raises the obvious question if similar methods could be derived to improve embeddings (i.e. semantic representations) of word sequences as well. We present a simpl ...

2017