Since Alan Turing published his paper on Computing Machinery and Intelligence, computers have come a long way. In a new paper published, scientists created a quantum brain that could revolutionize computing forever.
In today's modern age, where computer limitations are slowly fading, machine learning now has the ability to learn from its experiences.
Normally, this form of intelligence can only be achieved by using multiple computers and complex machine-learning algorithms. However, in a paper published in the journal Nature Nanotechnology on February 1st, scientists propose a new method. Designing a computer capable of embedded intelligence and taking advantage of the atoms' quantum spins may soon revolutionize computing.
The quantum brain is a prime example of neuromorphic computing. This form of computing recreates the neural structure of the human brain. As a step forward in the AI industry, neuromorphic computing enables robots embedded with small computing hardware in their 'brains' to someday make their own conscious decisions.
When you search for recipes, games, or your favorite restaurant, it may seem mundane, but on a wider scale, the calculations for each of your searches create substantial carbon footprints that keep growing every day.
Hence, Alexander Khajetoorians, the author of the study, and a professor from Radboud University have set out to find a way to meet computation's growing requirements while lowering energy-footprints.
In a statement, Bet Kappen, a co-author of the study and professor of Neural networks, says, "Until now, this technology, which is based on a century-old paradigm, worked sufficiently. But, it is a very energy-inefficient process."
Khajetoorians explains that their idea of building a 'quantum brain' is based on the quantum properties of materials that could be the basis for future solutions and applications in AI.
The Quantum Brain
The quantum brain isn't a steampunk version of ours. It is, however, hardware that tries to mimic the internal structure of human brains, including synapses and neurons.
Human brains use signals sent via neurons to make computations. Likewise, the quantum brain mimicks the brain's biological construction using cobalt atoms on a superconducting black phosphorus surface.
The atoms used have quantum properties: unique spin states. Researchers set out to see if it was possible to embed information on the atoms' spin states that will simulate 'neuron firing' between applied voltages.
The team was able to model biological neuron behaviors by using voltages to the network of atoms. They observed that the atoms demonstrated self-adaptive behavior based on what they remembered.
Khajetoorians says, "When stimulating the material over prolonged periods of time with voltage, we were surprised to see the synapses change. Where the material adapted its reaction based on external stimuli that it received. In short, it learned by itself."
Quantum Brains in Robots
Despite the discovery, quantum brains aren't coming to robots near you anytime soon. The authors explain that there are many unknown factors that need to be resolved within the system before its debut.
Check out more news and information on Tech & Innovation on Science Times.