Benoît Miramond
22
Documents
Présentation
Benoît Miramond est professeur au laboratoire LEAT de l'Université Côte d'Azur à Sophia-Antipolis. Auparavant, il était maître de conférence au laboratoire ETIS de l'Université de Cergy-Pontoise. Benoît Miramond a reçu l'Habilitation (HDR) en sciences de l'ingénieur et des systèmes en 2014 et le titre de doctorat en informatique en 2003.
Il est actuellement à la tête du groupe d'ingénierie neuromorphique eBRAIN du LEAT. Il dirige le projet de recherche suivants :
- projet ANR DeepSee (Deep Spiking Neural Networks for embedded systems) en collaboration avec les entreprises Renault, Prophesee et les laboratoires Cerco et I3S. Le projet se déroule de mars 2021 à septembre 2024.
- projet international ANR SOMA (Architecture de machine auto-organisée basée) en collaboration avec l'INRIA de Bordeaux, le LORIA de Nancy et l'HESSO de Genève de 2018 à 2022.
Dans ce contexte, ses recherches suivent une approche interdisciplinaire pour explorer de nouvelles architectures matérielles adaptatives et faible consommation inspirées des neurosciences et des sciences cognitives pour des applications d'Intelligence Artificielle embarquée.
Plus de détails sur [http://sites.unice.fr/site/bmiramond/Perso/](http://sites.unice.fr/site/bmiramond/Perso/ "Page web Miramond")
Benoît Miramond is Full Professor at LEAT Lab in the Université Côte d'Azur in Sophia-Antipolis. Previously, from 2005 to 2015, he was Associate Professor in the ETIS Lab in the University of Cergy-Pontoise. Benoît Miramond received the Habilitation thesis in sciences of engineering and systems in 2014 and the title of PhD in computer sciences in 2003.
He is currently the head of the eBRAIN research group in Neuromorphic Engineering from LEAT lab. He is leader of the following research projects :
- international ANR project SOMA (neural-based Self-Organizing Machine Architecture) in collaboration with INRIA at Bordeaux, LORIA at Nancy and HESSO in Geneva from 2018 to 2021.
- ANR DeepSee (Deep Spiking Neural Networks for Embedded and Autonomous systems) with Renault, Prophesee and the resarch labs Cerco and I3S. The project goes from march 2021 to september 2024.
In this context, his research is following an interdisciplinary approach to explore novel adaptive and low-power hardware architectures inspired from neurosciences and cognitive sciences for embedded AI applications.
More details: [http://sites.unice.fr/site/bmiramond/Perso/](http://sites.unice.fr/site/bmiramond/Perso/ "Web Page for Miramond Benoit")
Publications
- 7
- 3
- 3
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 22
- 22
- 3
- 1
Estimation of Energy Efficiency of Spiking Neural Networks on neuromorphic hardwareConférence INT, Iinstitut de Neurosciences de la Timone, Mar 2023, Marseille (13), France
Communication dans un congrès
hal-04044438v1
|
|
Cortex Inspired Learning to Recover Damaged Signal Modality by ReD-SOM ModelIJCNN 2023, IEEE, Jun 2023, Gold Coast, Australia, Australia. pp.01-09, ⟨10.1109/IJCNN54540.2023.10191701⟩
Communication dans un congrès
hal-04141988v1
|
|
Time series prediction and anomaly detection with recurrent spiking neural networksIJCNN 2023, IEEE, Jun 2023, Queensland, Australia. pp.10, ⟨10.1109/IJCNN54540.2023.10191614⟩
Communication dans un congrès
hal-04138502v1
|
|
An Analytical Estimation of Spiking Neural Networks Energy EfficiencyInternational Conference on Neural Information Processing ( ICONIP), Nov 2022, ITT Indore, India. pp.8, ⟨10.1007/978-3-031-30105-6_48⟩
Communication dans un congrès
hal-03875214v1
|
|
Self-Organizing Neurons: Toward Brain-Inspired Multimodal AssociationNeural Interfaces and Artificial Senses (NIAS), Sep 2021, Online, Spain. pp.1, ⟨10.29363/nanoge.nias.2021.007⟩
Communication dans un congrès
hal-03355250v1
|
|
Learning from Event Cameras with Sparse Spiking Convolutional Neural NetworksInternational Joint Conference On Neural Networks 2021 (IJCNN 2021), Jul 2021, Conférence virtuelle, China. pp.1-8
Communication dans un congrès
hal-03208690v1
|
|
Neuromorphic architectures, a support to third generation of Artificial Neural Networks and a new path toward low-power embedded AIMicro Innovation Day II, pôle SCS., Oct 2021, Marseille, France
Communication dans un congrès
hal-03435985v1
|
|
Improving Self-Organizing Maps with Unsupervised Feature ExtractionInternational Conference on Neural Information Processing (ICONIP) 2020, Aug 2020, Bangkok, Thailand. pp.474-486
Communication dans un congrès
hal-02934260v1
|
|
Neural coding: adapting spike generation for embedded hardware classificationIEEE World Congress on Computational Intelligence (WCCI) 2020, Jul 2020, Glasgow, United Kingdom. pp.8, ⟨10.1109/IJCNN48605.2020.9207702⟩
Communication dans un congrès
hal-02515109v1
|
|
GPU-based Self-Organizing-Maps for Post-Labeled Few-Shot Unsupervised LearningInternational Conference on Neural Information Processing (ICONIP) 2020, Aug 2020, Bangkok, Thailand. pp.404-416, ⟨10.1007/978-3-030-63833-7_34⟩
Communication dans un congrès
hal-02934233v1
|
|
An FPGA-based Hybrid Neural Network accelerator for embedded satellite image classificationIEEE International Symposium on Circuit and Systems (ISCAS 2020), May 2020, Seville, Spain. pp.5, ⟨10.1109/ISCAS45731.2020.9180625⟩
Communication dans un congrès
hal-02445183v1
|
|
Spike Nets for low power image processing4th Huawei Future ISP (Image Signal Processing) technologies workshop, Sep 2020, Sophia Antipolis, France
Communication dans un congrès
hal-02959675v1
|
Space Use-Case: Onboard Satellite Image Classificationspringer. Towards Ubiquitous, Low-power Image Processing Platforms, , pp.199-218, 2020, 978-3-030-53532-2. ⟨10.1007/978-3-030-53532-2_12⟩
Chapitre d'ouvrage
hal-02880903v1
|
Déclaration d’Invention DI-14851-01 du logiciel MicroAIFrance, N° de brevet: DI-14851-01. EDGE. 2021
Brevet
hal-03595177v1
|