Quantum Memristor combines AI and quantum computing

Artificial intelligence has developed more and more in recent years, with applications such as voice recognition, image identification, medical diagnosis and many others. Quantum technology, on the other hand, is capable of handling much more power than the most powerful supercomputer in the world.

Abstract representation of a neural network made up of photons and endowed with a memory capacity potentially linked to artificial intelligence. Image credit: Equinox Graphics, University of Vienna.

Scientists at the University of Vienna have now created a new device known as a quantum memristor that could bring these two worlds together, unlocking previously unimaginable powers. The experiment was carried out in partnership with the Italian National Research Council (CNR) and Politecnico di Milano on an embedded quantum processor that works on single photons. The research was published in the journal Nature Photonics.

Neural networks are mathematical models that are at the heart of all artificial intelligence applications. The biological structure of the human brain, made up of linked nodes, inspired these models.

Neural networks can be mathematically trained by adjusting their internal structure until they are subjected to human tasks, such as recognizing faces, interpreting medical images for diagnosis, and even tracking driving, much like how the brain learns by continually rearranging the connections between neurons.

Developing integrated devices capable of quickly and efficiently performing the necessary calculations in neural networks has therefore become an academic and industrial research priority.

One of the major game changers in the field was the development of the memristor in 2008. The memory-resistor, or memristor, is a device that adjusts its resistance based on a memory of a previous current. Experts noticed almost instantly that the unique behavior of memristors was remarkably comparable to that of neural synapses (among many other applications). As a result, the memristor has become a crucial element in neuromorphic designs.

Professor Philip Walther and Dr Roberto Osellame from the University of Vienna, the National Research Council (CNR) and the Politecnico di Milano have now suggested that it is possible to design a device that behaves like a memristor while operating on quantum states and also being able to encode and transmit quantum information. In other words, a quantum memristor.

The dynamics of a memristor seem to violate usual quantum behavior, which makes it difficult to realize such a device.

Physicists overcame the difficulty by employing single photons, or single quantum particles of light, and utilizing their unique ability to propagate continuously in a combination of two or more pathways. Single photons propagate through laser-etched waveguides on a glass surface and are directed over a superposition of many pathways in their experiment.

One of these pathways is used to sense the flux of photons passing through the device, and this quantity regulates the transmission on the other output via a complex electronic feedback mechanism, resulting in the desired memristive behavior.

Researchers have developed simulations indicating that optical networks containing quantum memristors can be used to learn about both classical and quantum tasks, implying that the quantum memristor could be the missing link linking artificial intelligence and science. quantum computing.

Unleashing the full potential of quantum resources within artificial intelligence is one of the biggest challenges in current research in quantum physics and computer science..

Michele Spagnolo, Study First Author, Faculty of Physics, Vienna Center for Quantum Science and Technology, University of Vienna

According to a recent study by Philip Walther of the University of Vienna, robots can learn faster when they use quantum resources and borrow strategies from quantum computing. This remarkable discovery is one step closer to a world in which quantum artificial intelligence is a reality.

Journal reference:

Spagnolo, M. et al. (2022) Experimental Photonic Quantum Memristor. Natural photonics. doi.org/10.1038/s41566-022-00973-5.

Source: https://www.univie.ac.at/fr/

Sherry J. Basler