Projects

UNITE: Brain-inspired device-circuits-algorithm co-design for resource-constrained hardware on the edge

UNITE's objective is to build a low-power few-shot learning on the edge, by co-desiging across device physics, circuits, architecture and algorithms. We combine the power of temporal computation using delays, with sparse connectivity, and few-shot learning algorithms for an extreme-edge self-adapting device. 

In METASPIN, we envision a radically new low-power artificial synapse technology based on spintronics nanodevices that will prevent catastrophic forgetting, i.e. the loss of memory of previously learned tasks upon learning new ones, a major flaw currently faced by all artificial intelligence applications.
We will develop a new class of neuromorphic hardware that will use magneto-ionics to support synaptic metaplasticity, i.e. a feature inspired by the human brain based on assigning a ‘hidden value’ to the states of artificial synapses to encode how important each state is. This will make it easier or harder to reconfigure the synaptic state upon learning a new task, giving a hierarchy to previously learned information and thus preventing catastrophic forgetting. The synaptic states will be given by the two magnetisation orientations in ferromagnets with perpendicular magnetic anisotropy, and by ferro/antiferromagnetic order in materials where the two phases coexist. In all cases, magneto-ionic gating will be used to locally modulate intrinsic magnetic properties to assign ‘hidden states’ to each synaptic state. The magneto- ionic hidden states will translate into a modulation of the switching probability between synaptic states, introducing the metaplasticity functionality. In parallel, we will develop ANNs learning schemes, adapted to our device physics and inspired by biological synaptic activity, that can learn with mitigated catastrophic forgetting. The ultimate goal of this project is to integrate this advanced synaptic technology and learning algorithms into an ANN demonstrator to test multitask learning on proof-of-concept tasks inspired by medical AI, and assess the impact of metaplasticity in catastrophic forgetting.

https://metaspin.c2n.universite-paris-saclay.fr/

• Master’s thesis on Online few-shot learning on small-world graphical hardware
• Master’s thesis on Exploiting RRAM-aware delays in SNNs for processing speech
• Master’s thesis on Evolving Optimized Neuromorphic Hardware Architectures
• Master’s thesis on Designing low-power and low-area sensing circuitry for resistive memory devices