The University of Gothenburg (UoG) has recently developed computers that can work as energy efficiently as human brains.
Indeed, alongside the Japan Tohoko University, researchers have been able to combine memory and calculation functions in the same component, hence successfully connecting oscillator networks and memristors. This project called the Topspin (Topotronic Multi-Dimensional Spin Hall Nano-Oscillator Networks) could help create more efficient technologies across the broad spectrum of mobile telephones, drones, and self-driving vehicles.
The UoG’s department of physics has also received a 150,000€ proof-of-concept grant from the European Research Council (ERC) in order to develop and commercialise advances for neuromorphic computing projects. With this, the project could lead the way towards more brain-like computers as well as develop a new, completely magnetic, quantum-inspired Ising machine based on two-dimensional networks of spin-Hall nano-oscillators.
This programme was able to go ahead following a sustained period of research and significant funding by state agencies in Sweden and Japan from 2019 to 2021. Tests are still ongoing in Gothenburg for the project as it is trying to find solutions to accelerate the development of energy-efficient and tunable artificial neurons able to emulate brain-inspired processes.