Zobrazeno 1 - 10
of 249
pro vyhledávání: '"Primeau, Francois"'
Autor:
Bensaid, Reda, Gripon, Vincent, Leduc-Primeau, François, Mauch, Lukas, Hacene, Ghouthi Boukli, Cardinaux, Fabien
In recent years, the rapid evolution of computer vision has seen the emergence of various foundation models, each tailored to specific data types and tasks. In this study, we explore the adaptation of these models for few-shot semantic segmentation.
Externí odkaz:
http://arxiv.org/abs/2401.11311
Deep learning (DL)-based solutions have emerged as promising candidates for beamforming in massive Multiple-Input Multiple-Output (mMIMO) systems. Nevertheless, it remains challenging to seamlessly adapt these solutions to practical deployment scenar
Externí odkaz:
http://arxiv.org/abs/2401.10513
Autor:
Hojatian, Hamed, Mlika, Zoubeir, Nadal, Jérémy, Frigon, Jean-François, Leduc-Primeau, François
Publikováno v:
IEEE Transactions on Machine Learning in Communications and Networking 2024
Hybrid beamforming (HBF) and antenna selection are promising techniques for improving the energy efficiency~(EE) of massive multiple-input multiple-output~(mMIMO) systems. However, the transmitter architecture may contain several parameters that need
Externí odkaz:
http://arxiv.org/abs/2308.06376
Autor:
Dupuis, Théo, Fournier, Yoan, AskariHemmat, MohammadHossein, Zarif, Nizar El, Leduc-Primeau, François, David, Jean Pierre, Savaria, Yvon
Convolutional Neural Networks (CNNs) are used in a wide range of applications, with full-precision CNNs achieving high accuracy at the expense of portability. Recent progress in quantization techniques has demonstrated that sub-byte Quantized Neural
Externí odkaz:
http://arxiv.org/abs/2306.09905
Autor:
AskariHemmat, MohammadHossein, Dupuis, Theo, Fournier, Yoan, Zarif, Nizar El, Cavalcante, Matheus, Perotti, Matteo, Gurkaynak, Frank, Benini, Luca, Leduc-Primeau, Francois, Savaria, Yvon, David, Jean-Pierre
In this paper, we present Quark, an integer RISC-V vector processor specifically tailored for sub-byte DNN inference. Quark is implemented in GlobalFoundries' 22FDX FD-SOI technology. It is designed on top of Ara, an open-source 64-bit RISC-V vector
Externí odkaz:
http://arxiv.org/abs/2302.05996
Energy-efficient deep neural network (DNN) accelerators are prone to non-idealities that degrade DNN performance at inference time. To mitigate such degradation, existing methods typically add perturbations to the DNN weights during training to simul
Externí odkaz:
http://arxiv.org/abs/2211.11561
Hybrid beamforming is a promising technology to improve the energy efficiency of massive MIMO systems. In particular, subarray hybrid beamforming can further decrease power consumption by reducing the number of phase-shifters. However, designing the
Externí odkaz:
http://arxiv.org/abs/2208.05443
Autor:
Kern, Jonathan, Henwood, Sébastien, Mordido, Gonçalo, Dupraz, Elsa, Aïssa-El-Bey, Abdeldjalil, Savaria, Yvon, Leduc-Primeau, François
Memristors enable the computation of matrix-vector multiplications (MVM) in memory and, therefore, show great potential in highly increasing the energy efficiency of deep neural network (DNN) inference accelerators. However, computations in memristor
Externí odkaz:
http://arxiv.org/abs/2205.01707
Autor:
Fu, Weiwei, Moore, J. Keith, Primeau, Francois, Collier, Nathan, Ogunro, Oluwaseun O., Hoffman, Forrest M., Randerson, James T.
The International Ocean Model Benchmarking (IOMB) software package is a new community resource used here to evaluate surface and upper ocean variables from CMIP5 and CMIP6 Earth System Models (ESMs) Our analysis reveals general improvement in the mul
Externí odkaz:
http://arxiv.org/abs/2202.12933
Autor:
Kern, Jonathan, Dupraz, Elsa, Aïssa-El-Bey, Abdeldjalil, Varshney, Lav R., Leduc-Primeau, François
This paper presents a quantized Kalman filter implemented using unreliable memories. We consider that both the quantization and the unreliable memories introduce errors in the computations, and develop an error propagation model that takes into accou
Externí odkaz:
http://arxiv.org/abs/2109.01520