This “quantum memristor” could enable brain-like quantum computers

Quantum and neuromorphic approaches both promise to fundamentally rewrite the way we run computers. And now they’ve been brought together after researchers developed a ‘quantum memristor’ that could form the basis of quantum neural networks.

During Moore’s Law seems to still have life in itthe borderThe limitations of conventional computing are becoming apparent, and there is growing interest in entirely different types of information processing that could overcome these obstacles.

One possibility is quantum computing, which uses the properties of quantum computers to achieve exponential computational speeds on some specific problems. Another possibility is to rewire our computer chips to more faithfully replicate how our brains work, known as neuromorphic computing.

The two approaches aim to improve very different aspects of conventional computing, and the synergies between the two are far from obvious. But that could change according to the researcherS demonstrated the first-ever neuromorphic component capable of processing quantum information.

The component in question is called a memristor, a Surname that comes from a combination of memory and resistance. These devices change their resistance based on how much current has passededit through them in the past, essentially storing a memory of their previous state.

This ability has caught the attention of neuromorphists because it mimics the behavior of biological synapses – the connections between neurons in the brain-who change the strength of their connections depending on how often they fire. There has been a range of recent research trying to use memristors to build more brain-like computers.

Physicists from the University of Vienna have now taken the idea one step further and developed a component that displays the same behavior when processing quantum information. The new device is described in a current paper in nature photonics.

Their so-called “quantum memristor” is based on integrated photonic technology that transports photons around a silicon chip to process information. But while photonic chips normally only perform classical calculations, the researchers designedit one that can manipulate the quantum states of the photons passing through.

To do this, use themedit the quantum principle of superposition: the idea that a quantum system can be in a combination of more than one state at the same time. They do this by giving the photon two paths and making it go through both at the same time.

This forms the basis for a qubit – the quantum equivalent of a bit – which can be used to encode information. Likewise, a bit can be either 0 or 1, the photon can be either in the first or second channel or, thanks to the strange properties of quantum mechanics, in a superposition of the two.

The researcherS’ The most important innovation, however, was to couple this system with additional circuitry that essentially counts the number of photons traveling through one of the paths and uses that to adjust the strength of the signal through the other path. The result is a device that can both process quantum information and exhibit memristive behavior.

To demonstrate their quantum memristor’s potential for practical computational tasks, they then created a computer model of the component and simulated what would happen if you string several together. They created a type of neural network based on a principle called reservoir computing, which essentially feeds data into a large network whose connections are fixed, and then trains just a single readout layer to interpret the output of that reservoir.

They showed that a system just off 3 Their quantum memristors learned to classify handwritten digits with 95 percent accuracy after training with just 1,000 images. The authors reported that achieve several classic reservoir computing schemesD poorer accuracies with significantly more computational resources and data.

The team also showed that a network of their devices could learn to perform quantum tasks beyond any classical device. They trained their device to detect whether quantum systems were entangled with an accuracy of 98 percent.

calculation figuring out how to use a device that combines two such different computing paradigms is going to take a lot of work. But quantum neural networks could be a powerful new tool in the era of Moore’s Law computing.

Photo credit: Equinox Graphics, University of Vienna


Are you looking for ways to stay ahead of the pace of change? Reconsider what is possible. Join a highly curated, exclusive cohort of 80 executives for Singularity’s flagship Executive Program (EP), a five-day, fully immersive executive transformation program that disrupts existing mindsets. Discover a new way of thinking, a new toolset and a network of fellow futurists dedicated to finding solutions to the rapidly changing world. Click here to learn more and apply today!

Leave a Reply

Your email address will not be published.