I am very happy to join Sony R&D in Zurich!
After five years working in the lab on modeling, training and benchmarking neuromorphic computing algorithms and systems, I think that we can push these solution into the real world.
To do that, we need to show our added value and competitive advantage at the system level. Therefore, we need to put together the pieces of the neuromorphic puzzle: sensors, processors and algorithms. And it starts from the sensors, where we trake inspiration from biology to only sense changes in the environment using evet-based vision sensors.
With the expansion of edge devices in our daily lives, the need for efficient and adaptive embedded systems is growing fast. However, the current Von Neumann architectures and AI models reach their technical and intellectual limits, and fail to address the current challenges of embedded intelligent systems. This is why we follow a brain-inspired computing approach based on the co-development of neuromorphic hardware and algorithms.
After three years of "online" interactions, it was great to meet again with these amazing people at MatNeC!
In the Zernike Institute Advent Calendar, we are presenting 24 short spotlights in December. In these specials, we highlight PhD students, postdocs, and technicians of our research groups - providing a glimpse into their typical day at work. In Episode 3 meet Dr. Lyes Khacef.
Lyes Khacef has received the first thesis prize from the STIC doctoral school in Université Côte d’Azur. The thesis discusses the modeled self-organizing artificial neural networks thanks to a mechanism of structural plasticity to create or cut connections, as well as a mechanism of synaptic plasticity which allows to modify the strength of these connections.
Le travail de Lyes a pour but d'explorer de nouvelles méthodes d'IA bio-inspirée. Il s'inspire du fonctionnement du cerveau humain, et plus précisément de la plasticité corticale, l'une de ses principales caractéristiques qui nous permet d'apprendre et de nous adapter.