Artificial Intelligence for Magnetic Devices in Quantum and Neuromorphic Computing

In the coming decades, a deep change in computational resources will occur due to the emergence of new technologies. This will provide a clear overall benefit to society because these computers will substantially improve the energetic efficiency and computational capacity on current technologies. There are two particularly important new technologies, one based on quantum technology using qubits as the basic unit. Qubit states are a coherent superposition (a linear combination) of the states |0> and |1>. This type of technology is under development with qubits using different supports, among others superconductors, ion-trapped systems and magnetic systems (Science, 2016, 354, 6316). Applications are quite specific in fields such as cryptography, quantum communication, quantum systems modeling and artificial intelligence. A second technology is based on the development of neuromorphic circuits that seek to emulate the energy-efficient information processing of the brain. Neuromorphic circuits, usually called memristors, aim to take advantage of the parallel, low-power, and fault-tolerant processing of the brain by mimicking the functions of neurons and synapses in highly interconnected networks (Nature Nanotechnology, 2020, 15, 517). Like quantum computing, research on neuromorphic devices is also at an initial stage of development. But early prototypes show enormous potential compared to classical computing, especially in artificial intelligence-related processes such as learning, problem-solving and decision-making by performing numerous tasks simultaneously.

In this project, the aim is to design and improve the efficiency of magnetic systems that are the basic units of both, qubits and memristors. We plan to take advantage of the synergy between two groups involved in this proposal with very different profiles, the UB group with both theoretical and experimental research in magnetic systems and the UPC group with a long experience in programming, signal processing and use of methods based on artificial intelligence.