Functionalised dense spintronics oscillator networks for neuromorphic computing

This project proposes the functionalisation of dense arrays of spintronic nanodevices called magnetic tunnel junctions (MTJs) for neuromorphic computing application.

Neuromorphic computing is a field that proposes a novel computing architecture aimed at reducing energy consumption and enhancing computational capabilities compared to conventional computers. It mimics the Human brain where memory and information processing are physically intertwined.

One key component is the synapse, which plays a crucial role in enabling the communication between neurons and serving as a dynamic, plastic memory where the strength of connections between neurons is stored.

This characteristic allows for tuning these connections in a non-volatile and reversible manner, forming the foundation for intricate learning and memory processes.For this purpose, the field of spintronics will be employed, focusing on controlling electron spin.

Spintronics has gained significant importance in data storage applications, and spintronics nanodevices such as MTJs have recently emerged as promising candidates for neuromorphic computing due to their robustness, multifunctionalities and compatibility with compatibility to metal-oxide-semiconductor (CMOS) technology.

They hold strong appeal for the development of artificial synapses and neurons that mimic their biological counterparts due to their low power consumption and fast switching compared to traditional transistors. In addition, the nanoscale size of spintronic devices allows for the creation of large neural networks.

The integration of memory and processing in these devices has the potential to revolutionize computational systems and strongly reduce energy consumption while increasing computational speed.