Signals to Spikes for Neuromorphic Regulated Reservoir Computing and EMG Hand Gesture Recognition

Archive ouverte : Communication dans un congrès

Garg, N. | Balafrej, I. | Beilliard, Y. | Drouin, Dominique | Alibart, F. | Rouat, J.

Edité par HAL CCSD ; Association for Computing Machinery

International audience. Surface electromyogram (sEMG) signals result from muscle movement and hence they are an ideal candidate for benchmarking event-driven sensing and computing. We propose a simple yet novel approach for optimizing the spike encoding algorithm's hyper-parameters inspired by the readout layer concept in reservoir computing. Using a simple machine learning algorithm after spike encoding, we report performance higher than the state-of-The-Art spiking neural networks on two open-source datasets for hand gesture recognition. The spike encoded data is processed through a spiking reservoir with a biologically inspired topology and neuron model. When trained with the unsupervised activity regulation CRITICAL algorithm to operate at the edge of chaos, the reservoir yields better performance than state-of-The-Art convolutional neural networks. The reservoir performance with regulated activity was found to be 89.72% for the Roshambo EMG dataset and 70.6% for the EMG subset of sensor fusion dataset. Therefore, the biologically-inspired computing paradigm, which is known for being power efficient, also proves to have a great potential when compared with conventional AI algorithms. © 2021 Owner/Author.

Consulter en ligne

Suggestions

Du même auteur

Oxygen vacancy engineering of TaO x -based resistive memories by Zr doping ...

Archive ouverte: Article de revue

Palhares, João | 2021-07-12

International audience. Resistive switching (RS) devices are promising forms of non-volatile memory. However, one of the biggest challenges for RS memory applications is the device-to-device (D2D) variability, which...

Fully CMOS-compatible passive TiO2-based memristor crossbars for in-memory ...

Archive ouverte: Article de revue

El Mesoudy, Abdelouadoud | 2022-02

International audience. Brain-inspired computing and neuromorphic hardware are promising approaches that offer great potential to overcome limitations faced by current computing paradigms based on traditional von-Ne...

Voltage-dependent synaptic plasticity: Unsupervised probabilistic Hebbian p...

Archive ouverte: Article de revue

Garg, Nikhil | 2022-10-21

International audience. This study proposes voltage-dependent-synaptic plasticity (VDSP), a novel brain-inspired unsupervised local learning rule for the online implementation of Hebb’s plasticity mechanism on neuro...

Du même sujet

The new laws of outer space : ethics, legislation, and governance in the ag...

Livre | Pagallo, Ugo

This text maps out the moral, legal and societal issues brought forth by the use of autonomous systems such as AI and smart robots in outer space. Humanity is on the brink of a new space era in which projects for permanent human c...

L' intelligence artificielle n'existe pas / Luc Julia

Livre | Julia, Luc (1966-....). Auteur | 2019

"Vous ne comprenez rien à l'intelligence artificielle (IA) ? Vous avez peur que de méchants robots prennent le pouvoir et finissent par contrôler le monde ? Vous vous intéressez à l'IA et aux nouvelles technologies, mais vous aime...

Manager l'intelligence artificielle / Roland Robeveille, Michelle Veyssière

Livre | Robeveille, Roland (19..-....) - docteur en gestion. Auteur | 2021 - 2e édition

Les machines intelligentes - Intelligence Artificielle ou IA - font émerger une complexité nouvelle dans le monde professionnel et dans nos vies privées. L'intrusion d'internet, du numérique et de l'IA est telle que les rôles et m...

Intelligence artificielle et innovations digitales en santé : transformatio...

Livre | Chassagne, Valentin. Auteur | 2021

"Depuis cinq ans environ, la donnée de santé, exploitée grâce à l’intelligence artificielle et au digital, entraine une transformation profonde du secteur de la santé, qui révolutionne à la fois les modes de prise en charge, les m...

I.A. la plus grande mutation de l'histoire : Qui dominera l 'I.A. dominera ...

Livre | Lee, Kai-Fu (1961-....). Auteur | 2021

La 4ème de couv. indique : "L'intelligence artificielle est la plus grande révolution de l'humanité. Bien plus brutale car plus rapide que la Révolution industrielle, elle pourrait rebattre les cartes de l'ordre mondial. Qui gagne...

Five simultaneous artificial intelligence data challenges on ultrasound, CT...

Archive ouverte: Article de revue

Lassau, N. | 2019-04

International audience. PurposeThe goal of this data challenge was to create a structured dynamic with the following objectives: (1) teach radiologists the new rules of General Data Protection Regulation (GDPR), whi...

Chargement des enrichissements...