top of page
  • X
  • LinkedIn
  • Youtube

SkyANN

Skyrmionic  Artificial  Neural  Network


About the project

 

SkyANN presents a groundbreaking paradigm for neuromorphic computing. We closely emulate the neurophysiology of the brain by combining skyrmionic quasiparticles with electrical CMOS connections. The skyrmions mimic neurotransmitters and facilitate complex computations at the synapse level, while electrical CMOS connections simulate the propagation of action potentials among neurons for rapid and dense inter-layer connectivity.

Targeting low-power neuromorphic computing, our innovative magneto-electric devices aim to achieve an energy consumption that is four orders of magnitude lower than the current CMOS technology while doubling the bandwidth for the same device footprint. This will enhance edge inference and learning capabilities. This approach challenges contemporary neural networks implemented with CMOS digital, mixed-signal, and emerging in-memory computing technologies, which are limited by lower energy efficiency and reliability.

 

Consortium

The SkyANN consortium is carefully balanced, its members span all the necessary technical competences and fields of research for this project and have access to infrastructure that exceeds standards. The project is built on leading expertise and strong prior collaborative links between partners. This long-standing collaboration has led to groundbreaking research and key publications that are foundational for SkyANN.

Partners cover the entire value chain of SkyANN starting from physics to the design of novel computational systems, from materials to devices to circuits and CMOS interfacing all the way to prototyping, evaluation and validation of technologies and their exploitation for future commercial cases.

SkyANN is therefore uniquely positioned to enable technical breakthroughs and innovation in the space of artificial neural networks and deliver ultra-energy-efficient intelligence at the edge, significantly reducing the carbon emissions associated with AI and enhancing data privacy.

The consortium also includes a strong private sector component with the involvement of Thales and Eurida, ensuring that the results will be exploited and live on after the project ends.

bottom of page