Menu
Close
Sign up for NPL updates
Close
Sign up for NPL updates

For people, place, prosperity and planet, we deliver impact with measurement science.

  • Home
  • News
  • Artificial networks for future computing

New insights into novel artificial networks for future computing.

Artificial intelligence applications are experiencing a boom and expected to be mainstream technologies in the near future. However, these applications are running on classic computing hardware and are extremely power hungry

2 minute read

New research published in Nature: Communications Materials.

Artificial intelligence applications are experiencing a boom and expected to be mainstream technologies in the near future. However, these applications are running on classic computing hardware and are extremely power hungry.  

This creates opportunities for development of new energy-efficient hardware solutions inspired by nature, e.g. brain-like computing. Some of the best-known examples are neuromorphic computing and neural networks  which mimics the way the human brain works. One of the possible realisations of such neural networks is through an Artificial Spin Ice (ASI) lattice. NPL and partners* have investigated the impact of introducing hexagonal magnetic defects into such ASI structure.  

Through the interdisciplinary research, the international team successfully demonstrated a mechanism for tailoring the ASI system’s behaviour introducing the designed magnetic defects causing stochastic topological excitations in the system and controlling the dynamics of the ASI-based neural networks. The implication of this discovery is expected to be used in applications including magnetic memory devices and spin-based logic applications.  

The results of this study provide insights into the collective and stochastically-controlled behaviour in artificial neural networks realised through the magnetic ASI lattice and pave the way for future research into for such emerging applications as reconfigurable spin-waveguides and hardware realisations of low-energy future computing systems.  

Olga Kazakova, NPL Fellow, said: “This work demonstrates a very important milestone for us. Being able to controllably create topological states associated with ASI defects and demonstrate stochastic but statistically predictable behaviours within the ASI lattice. The results bring us closer to realisation of energy-efficient neuromorphic computing. This was an outcome of a great international collaboration with large research facilities in the UK, Germany and France. 

*partners include: Centre d’Elaboration de Materiaux et d’Etudes Structurales, France, School of Electronic and Electrical Engineering, University of Leeds, UK, Physikalisch-Technische Bundesanstalt, Germany and Université Paul Sabatier, Université de Toulouse, France.  

Find out more here

03 Oct 2024