BILAVARN , Sébastien

Next level of energy efficiency is an essential concern in the evolution of semiconductor technologies sincethe widespread development of distributed, nomad and embedded systems. The best example is that ofmobile phones the last twenty years but consumer electronics perspectives (IoT, 5G, HPC, ArtificialIntelligence, Robotics, automotive...

National audience

A significant factor in the evolution of modern electronic systems is the appearance of new architectures based on the programming of hardware components such as Field programmable Gate Arrays (FPGAs). The introduction of those components as an alternative computation unit and the flexibility they offer increase the interest of suche an...

Efficient acceleration of deep convolutional neural networks is currently a major focus in Edge Computing research. This paper presents a realistic case study on ResNet-18, exploring Partial Reconfiguration (PR) as an alternative to the standard static reconfigurable approach. The PR strategy is based on sequencing the layers of the DNN on a...

L''intelligence artificielle (IA) est devenue très populaire dans de nombreuses applications embarquées (Edge Computing). Mais l'exécution d'algorithmes d'IA sous contraintes de ressources matérielles limitées présentent de nombreux défis en termes de complexité, de contraintes mémoire, de performance et encore plus concernant l'efficacité...

Efficient acceleration of deep convolutional neural networks is currently a major focus in Edge Computing research. This paper presents a realistic case study on ResNet-18, exploring Partial Reconfiguration (PR) as an alternative to the standard static reconfigurable approach. The PR strategy is based on sequencing the layers of the DNN on a...

With the increasing demand for edge AI application scenarios, as the most popular deep learning models, Convolutional Neural Networks (CNNs) need advanced solutions for the deployment of highly energy-efficient implementations. This paper presents a novel approach to improve the efficiency of CNN inference on Field-Programmable Gate Arrays...

Emerging applications in the field of multimedia like new video standards require the capabilities of multiprocessor architectures which are a high source of heat dissipation. In this chapter, we present the definition of a power model suited for multiprocessor power management based on the study of a H264/AVC decoder implementation. We will...

Efficient acceleration of deep convolutional neural networks is currently a major focus in Edge Computing research. This paper presents a realistic case study on ResNet-18, exploring Partial Reconfiguration (PR) as an alternative to the standard static reconfigurable approach. The PR strategy is based on sequencing the layers of the DNN on a...

National audience

National audience