MAGNONIC ARTIFICIAL NEURAL NETWORKS AND GATE ARRAYS

We seek to explore and challenge the limits of spin-based devices and their energy efficiency. This will be achieved by combining two inherently energy-efficient technology paradigms: (i) magnonics (using spin waves – low energy magnetic excitations – to process signals and data) and (ii) neuromorphic computing (using large-scale integrated systems and analog circuits to solve data-driven problems in a brain-like manner). We will use nanoscale chiral magnonic resonators as building blocks of artificial neural networks. The power of the networks will be demonstrated by creating magnonics versions of field programmable gate arrays, reservoir computers, and recurrent neural networks. The ultimate efficiency of the devices will be achieved by (a) maximising their magnetic nonlinearity; (b) using epitaxial yttrium iron garnet, which has the lowest known magnetic damping, for thin film magnonic media and resonators; and (c) using wireless delivery of power. Sensitive to the resonators’ micromagnetic states, such artificial neural networks will be conveniently programmable and trainable within existing paradigms of magnetic data storage. The latter includes magnetic random-access memory (MRAM), which is already compatible with CMOS, while compatibility with other technology paradigms of spintronics will also be sought, explored, and exploited. Thereby, the key ambition of our proposed very forward-looking research programme is to develop and establish a novel, revolutionary class of energy-efficient spin-based components and devices for use in green high-tech data communication, processing, and storage technologies, thereby helping unlock the full potential of spintronics. We will seek dissemination of our developed and appropriately protected designs, processes, and technologies to interested European and international companies, thereby improving the competitiveness of the European high-tech industry.