J
Jean-Michel Portal
Researcher at Centre national de la recherche scientifique
Publications - 145
Citations - 2335
Jean-Michel Portal is an academic researcher from Centre national de la recherche scientifique. The author has contributed to research in topics: Resistive random-access memory & Artificial neural network. The author has an hindex of 25, co-authored 136 publications receiving 2047 citations. Previous affiliations of Jean-Michel Portal include Alternatives & Aix-Marseille University.
Papers
More filters
Journal ArticleDOI
Non-Volatile Flip-Flop Based on Unipolar ReRAM for Power-Down Applications
TL;DR: The architecture of the RNVFF, based on the insertion of a non-volatile memory block before a master-slave Flip-Flop, is detailed and is fully validated through electrical simulations.
Proceedings ArticleDOI
In-memory resistive RAM implementation of binarized neural networks for medical applications
Bogdan Penkovsky,Marc Bocquet,Tifenn Hirtzlin,Jacques-Olivier Klein,Etienne Nowak,Elisa Vianello,Jean-Michel Portal,Damien Querlioz +7 more
TL;DR: This work suggests strategies to apply BNNs to biomedical signals such as electrocardiography and electroencephalography, keeping accuracy level and reducing memory requirements, and investigates the memory-accuracy trade-off when binarized whole network and binarizing solely the classifier part.
Posted Content
Implementing Binarized Neural Networks with Magnetoresistive RAM without Error Correction
Tifenn Hirtzlin,Bogdan Penkovsky,Jacques-Olivier Klein,Nicolas Locatelli,Adrien F. Vincent,Marc Bocquet,Jean-Michel Portal,Damien Querlioz +7 more
TL;DR: In this paper, the authors show that bit errors can be tolerated by binarized neural networks to an outstanding level, based on examples of image recognition tasks (MNIST, CIFAR-10 and ImageNet): bit error rates of ST-MRAM up to 0.1% have little impact on recognition accuracy.
Journal ArticleDOI
Dégradation de l'isoproturon et disponibilité de ses résidus dans le sol
TL;DR: In this article, the degradation half-life of extractable isoproturon was an estimated 14 days, and 72% of the radiolabel added remained in the soil as non-extractable residue.
Proceedings ArticleDOI
Resistive memories: which applications?
Fabien Clermidy,N. Jovanovic,Santhosh Onkaraiah,Houcine Oucheikh,Olivier Thomas,Ogun Turkyilmaz,Elisa Vianello,Jean-Michel Portal,Marc Bocquet +8 more
TL;DR: The first application is FPGA, one of the first architecture that can benefit the most from ReRAM integration to reduce footprint and save energy, and the second application relates to ultra-low-power systems and the way to obtain an instantaneous "freeze" mode in devices for Internet of Things.