معرفی
Jean Tomas is an Associate Professor at the University of Bordeaux, affiliated with the Laboratoire de l'intégration, du matériau au système (IMS). He is a member of the Bioelectronics research group and Team 2HC within IMS, a prominent research laboratory focused on materials, systems integration, and micro/nanoelectronics. His work bridges the gap between theoretical neural models and practical hardware implementations.
Dr. Tomas's research focuses on neuromorphic engineering with particular emphasis on spiking neural networks and memristive systems. His work spans from fundamental circuit design of biomimetic neurons to practical applications in low-power computing and intelligent sensors. He has made significant contributions to the hardware implementation of spiking neural networks using memristive technologies, exploring how these systems can achieve efficient computation with minimal energy consumption. His research also extends to electronic microassembly processes and power electronics.
The analysis of his publication record reveals a consistent trajectory from early work on analog neural circuit design toward increasingly sophisticated neuromorphic systems. His recent publications demonstrate expertise in mixed-mode spiking neural networks, passive memristive arrays, and their application to neuromorphic cameras. His work addresses critical challenges in hardware neural networks, including the effects of line resistance in crossbar arrays and strategies for on-the-fly learning in resource-constrained environments.
Dr. Tomas actively collaborates with researchers across multiple institutions, including CNRS@CREATE Ltd. and various university laboratories. His work has been supported by significant research initiatives including ANR-19-CHIA-0003 (GrAI - Green Artificial Intelligence) and the European Project ULPEC (Ultra-Low Power Event-Based Camera).
He is deeply involved in the Bioelectronics research ecosystem at IMS, contributing to both fundamental research and practical applications. His work connects circuit design with system-level implementation, particularly in the development of energy-efficient neuromorphic sensors that can process information at the edge of computing networks.
