Publications associées à Lecture performée

Dynamic Model Pruning with Feedback

Martin Jaggi, Luis Felipe Barba Flores, Sebastian Urban Stich, Tao Lin, Daniil Dmitriev

Deep neural networks often have millions of parameters. This can hinder their deployment to low-end devices, not only due to high memory requirements but also because of increased latency at inference. We propose a novel model compression method that gener ...

2020

The zaojing: review of a unique wooden construction typology

Corentin Jean Dominique Fivet, Jingxian Ye

The zaojing is an ancient Chinese construction system consisting of a dome-shaped assembly of interlocked wooden pieces. These structures are placed above live performance spaces to enhance sound diffusion while creating background scenery. Existing zaojin ...

2020

Informative sample generation using class aware generative adversarial networks for classification of chest Xrays

Jean-Philippe Thiran

Training robust deep learning (DL) systems for disease detection from medical images is challenging due to limited images covering different disease types and severity. The problem is especially acute, where there is a severe class imbalance. We propose an ...

2019

Negative Capacitance as Performance Booster for Tunnel FETs and MOSFETs: An Experimental Study

Mihai Adrian Ionescu, Christian Enz, Igor Stolichnov, Farzan Jazaeri, Ali Saeidi, Francesco Bellando

This letter reports for the first time a full experimental study of performance boosting of tunnel FETs (TFETs) and MOSFETs by negative capacitance (NC) effect. We discuss the importance of capacitance matching between a ferroelectric NC and a device capac ...

Institute of Electrical and Electronics Engineers2017

TILDE: A Temporally Invariant Learned DEtector

Pascal Fua, Vincent Lepetit, Kwang Moo Yi, Yannick Verdie

We introduce a learning-based approach to detect repeatable keypoints under drastic imaging changes of weather and lighting conditions to which state-of-the-art keypoint detectors are surprisingly sensitive. We first identify good keypoint candidates in mu ...

2015

Privacy-Sensitive Audio Features for Conversational Speech Processing

Sree Hari Krishnan Parthasarathi

The work described in this thesis takes place in the context of capturing real-life audio for the analysis of spontaneous social interactions. Towards this goal, we wish to capture conversational and ambient sounds using portable audio recorders. Analysis ...

EPFL2011

Métamatériau Mécano-Acoustique basé sur le Concept de Ligne de Transmission Duale

Hervé Lissek, Frédéric Bongard

Les métamatériaux acoustiques sont des structures artificielles conçues pour réaliser des propriétés macroscopiques inédites dans la nature, telles qu’un indice de réfraction négatif. Parmi les différentes configurations permettant d’obtenir de telles prop ...

2010

Innovative Manufacturing Technology for RF Passive Devices Combining Electroforming and CFRP Application

Michael Mattes

In this paper an innovative manufacturing technology is presented that combines electroforming and CFRP application and thus could be an enabling technology to achieve tuningless lightweight RF passive hardware with complex geometries. In order to demonstr ...

2008

A Primary-Backup Protocol for In-Memory Database Replication

Rodrigo Malta Schmidt, Fernando Pedone

The paper presents a primary-backup protocol to manage replicated in-memory database systems (IMDBs). The protocol exploits two features of IMDBs: coarse-grain concurrency control and deferred disk writes. Primary crashes are quickly detected by backups an ...

2006