Publications associées à Fonction de base radiale

A new variable shape parameter strategy for RBF approximation using neural networks

The choice of the shape parameter highly effects the behaviour of radial basis function (RBF) approximations, as it needs to be selected to balance between the ill-conditioning of the interpolation matrix and high accuracy. In this paper, we demonstrate ho ...

PERGAMON-ELSEVIER SCIENCE LTD2023

A Theory of Finite-Width Neural Networks: Generalization, Scaling Laws, and the Loss Landscape

Berfin Simsek

Deep learning has achieved remarkable success in various challenging tasks such as generating images from natural language or engaging in lengthy conversations with humans. The success in practice stems from the ability to successfully train massive neural ...

EPFL2023

Proximal Point Imitation Learning

Volkan Cevher, Luca Viano, Igor Krawczuk, Angeliki Kamoutsi

This work develops new algorithms with rigorous efficiency guarantees for infinite horizon imitation learning (IL) with linear function approximation without restrictive coherence assumptions. We begin with the minimax formulation of the problem and then o ...

2022

Learning of Continuous and Piecewise-Linear Functions With Hessian Total-Variation Regularization

Michaël Unser, Shayan Aziznejad, Joaquim Gonçalves Garcia Barreto Campos

We develop a novel 2D functional learning framework that employs a sparsity-promoting regularization based on second-order derivatives. Motivated by the nature of the regularizer, we restrict the search space to the span of piecewise-linear box splines shi ...

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC2022

Selective attention involves a feature-specific sequential release from inhibitory gating

David Pascucci, Gijs Plomp

Selective attention is a fundamental cognitive mechanism that allows our brain to preferentially process relevant sensory information, while filtering out distracting information. Attention is thought to flexibly gate the communication of irrelevant inform ...

2022

Theory of Deep Learning: Neural Tangent Kernel and Beyond

Arthur Ulysse Jacot-Guillarmod

In the recent years, Deep Neural Networks (DNNs) have managed to succeed at tasks that previously appeared impossible, such as human-level object recognition, text synthesis, translation, playing games and many more. In spite of these major achievements, o ...

EPFL2022

Multiscale Representation Learning of Graph Data With Node Affinity

Pascal Frossard, Chenglin Li, Xing Gao

Graph neural networks have emerged as a popular and powerful tool for learning hierarchical representation of graph data. In complement to graph convolution operators, graph pooling is crucial for extracting hierarchical representation of data in graph neu ...

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC2021

Evidence accumulation relates to perceptual consciousness and monitoring

Olaf Blanke, Fosco Bernasconi, Nathan Quentin Faivre, Mi Xue Tan, Michael Eric Anthony Pereira, Wenwen Chang

A fundamental scientific question concerns the neural basis of perceptual consciousness and perceptual monitoring resulting from the processing of sensory events. Although recent studies identified neurons reflecting stimulus visibility, their functional r ...

Nature Research2021

On evaluation of thermophysical properties of transformer oil-based nanofluids: A comprehensive modeling and experimental study

Farzaneh Talebkeikhah, Mohsen Talebkeikhah

Transformer oil-based nanofluids are known to have higher thermal conductivity and heat transfer performance compared to conventional transformer oils. In this study, four different types of transformer oil-based nanofluids are synthesized using the well-k ...

ELSEVIER2020

Investigation of effective processes parameters on lead (II) adsorption from wastewater by biochar in mild air oxidation pyrolysis process

Farzaneh Talebkeikhah, Mohsen Talebkeikhah, Mohammad Torkashvand

This study evaluates and compares several machine learning methods on the effects of different parameters in lead adsorption capacity. pH, contact time, adsorbent dosage and initial lead concentration were considered as inputs and adsorption capacity was r ...

TAYLOR & FRANCIS LTD2022

Learning Activation Functions in Deep (Spline) Neural Networks

Michaël Unser, Pakshal Narendra Bohra, Shayan Aziznejad, Harshit Gupta, Joaquim Gonçalves Garcia Barreto Campos

We develop an efficient computational solution to train deep neural networks (DNN) with free-form activation functions. To make the problem well-posed, we augment the cost functional of the DNN by adding an appropriate shape regularization: the sum of the ...

2020

Disentangling feature and lazy training in deep neural networks

Matthieu Wyart, Mario Geiger, Stefano Spigler, Arthur Ulysse Jacot-Guillarmod, Arthur Jacot

Two distinct limits for deep learning have been derived as the network width h -> infinity, depending on how the weights of the last layer scale with h. In the neural tangent Kernel (NTK) limit, the dynamics becomes linear in the weights and is described b ...

IOP PUBLISHING LTD2020

Muscle activities in similar arms performing identical tasks reveal the neural basis of muscle synergies

Martina Coscia

Are the muscle synergies extracted from multiple electromyographic signals an expression of neural information processing, or rather a by-product of mechanical and task constraints? To address this question, we asked 41 right-handed adults to perform a var ...

SPRINGER2020

Physico-chemical properties prediction of hydrochar in macroalgae Sargassum horneri hydrothermal carbonisation

Farzaneh Talebkeikhah, Mohsen Talebkeikhah

This study evaluates and compares several machine learning methods on the effects of different parameters in the hydrothermal carbonisation (HTC) process of macroalgae Sargassum horneri. Reaction temperature, residence time, biomass particle size, the amou ...

TAYLOR & FRANCIS LTD2021

Automated Synthesis of Adversarial Workloads for Network Functions

Rishabh Ramesh Iyer, Aikaterini Argyraki, Arseniy Zaostrovnykh, Jonas Fietz, Luis David Figueiredo Mascarenhas Moreira Pedrosa

Software network functions promise to simplify the deployment of network services and reduce network operation cost. However, they face the challenge of unpredictable performance. Given this performance variability, it is imperative that during deployment, ...

ACM2018

Neural Tanget Kernel: Convergence and Generalization in Neural Networks

Clément Hongler, Franck Raymond Gabriel, Arthur Ulysse Jacot-Guillarmod

At initialization, artificial neural networks (ANNs) are equivalent to Gaussian processes in the infinite-width limit [12, 9], thus connecting them to kernel methods. We prove that the evolution of an ANN during training can also be described by a kernel: ...

NEURAL INFORMATION PROCESSING SYSTEMS (NIPS)2018

Non-Linear Digital Self-Interference Cancellation for In-Band Full-Duplex Radios Using Neural Networks

Alexios Konstantinos Balatsoukas Stimming

Full-duplex systems require very strong self-interference cancellation in order to operate correctly and a significant part of the self-interference signal is due to non-linear effects created by various transceiver impairments. As such, linear cancellatio ...

IEEE2018

A Formally Verified NAT

George Candea, Solal Vincenzo Pirelli, Aikaterini Argyraki, Arseniy Zaostrovnykh, Luis David Figueiredo Mascarenhas Moreira Pedrosa

We present a Network Address Translator (NAT) written in C and proven to be semantically correct according to RFC 3022, as well as crash-free and memory-safe. There exists a lot of recent work on network verification, but it mostly assumes models of networ ...

Assoc Computing Machinery2017

On Structured Sparsity of Phonological Posteriors for Linguistic Parsing

Hervé Bourlard, Afsaneh Asaei, Milos Cernak

The speech signal conveys information on different time scales from short (20--40 ms) time scale or segmental, associated to phonological and phonetic information to long (150--250 ms) time scale or supra segmental, associated to syllabic and prosodic info ...

2016

Spline- and Wavelet-based Models of Neural Activity in Response to Natural Visual Stimulation

Joao Emanuel Felipe Gerhard, Luca Szegletes

We present a comparative study of the performance of different basis functions for the nonparametric modeling of neural activity in response to natural stimuli. Based on naturalistic video sequences, a generative model of neural activity was created using ...

Ieee2012