Breakthrough in the development of a new low-cost computer

(Press-News.org) A low-energy challenger to the quantum computer that also works at room temperature may be the result of research at the University of Gothenburg. The researchers have shown that information can be transmitted using magnetic wave motion in complex networks.

Spintronics explores magnetic phenomena in nano-thin layers of magnetic materials that are exposed to magnetic fields, electric currents and voltages. These external stimuli can also create spin waves, ripples in a material's magnetisation that travel with a specific phase and energy.

The researchers can generate and control the spin waves, enabling a phase controlled mutual synchronization between two so-called spin Hall nano-oscillators. By controlling the phase of these waves, the research team was able to generate binary phases across the network. For the first time, they showed that spin waves can mediate both in phase and out of phase between the oscillators. The phenomenon can be tuned by adjusting either the magnetic field, the electric current, the applied gate voltage or the distance between the oscillators.

Produces the best guess

These advances pave the way for the next generation of Ising machines, an alternative to quantum computers that require much less energy and can operate at room temperature.

Quantum computers and Ising machines are useful for solving so-called combinatorial optimisation problems, where the aim is to produce the best guess rather than the exact answer to a problem. Many AI models aim to produce guesses that are good enough for the purpose. In today's computers, these AI calculations require a lot of computing power and thus consume energy.

Networks of oscillators

“With the help of spin waves, we are closer to creating highly efficient, low-power computing systems that can solve real-world problems,” says Akash Kumar, lead author of the study published in the scientific journal Nature Physics.

Following the breakthrough, researchers at the University of Gothenburg are now building networks of hundreds of thousands of oscillators to develop the next generation of Ising machines. Because the oscillators operate at room temperature and have a nanoscale footprint, these devices can be easily adapted to larger systems, but also to smaller devices, such as a mobile phone.

“Spintronics has the potential to impact many different fields, from artificial intelligence and machine learning to telecommunications and financial systems. The ability to control and manipulate spin waves at the nanoscale could lead to the development of more powerful and efficient sensors, and even high-frequency stock trading machines,” says Akash Kumar.

Facts: Ising machine

An Ising machine is a new type of computational system that mimics how magnetic spins in a physical material organise themselves to reach a stable state. It is mainly used to solve complex optimisation problems in an efficient way. Rather than calculating step-by-step as in conventional computers, the system's many small ‘spins’ work together to quickly find the best solution. The Ising machine is programmed by the strength of the connections between the different spins. If the coupling is positive, the spins will point in the same direction (in phase) and if it is negative, they will point in the opposite direction (out of phase). The solution to the problem is then read out as the final direction of all the different spins after they have been optimally aligned.

END