Boosting neuromorphic computing, digital platforms and services
Collaborating universities in Sweden and Japan have made a breakthrough in the ongoing quest to make computers more energy efficient by making them work more like the human brain. Scientists from the University of Gothenburg and Tohoku University, located in the city of Sendai, 380 kilometers north of Tokyo, did so under the umbrella of the Topotronic Multi-Dimensional Spin Hall Nano research project. -Oscillator Networks, which is usually rather faster called “Topspin” (like in tennis or snooker).
Topspin’s goal is to be a catalyst to help create more efficient technologies in areas such as mobile handsets, satellites and autonomous vehicles. In this case, the fruit of the academic collaboration proves, for the first time, that it is possible to join oscillators and memristors, by merging them into a single unit combining both memory and calculation functions.
Recently, research interest in using coupled oscillator networks to improve computer science has increased dramatically due to the belief that oscillator-based systems may well be developed faster and more than digital circuits. traditional. Meanwhile, a memristor limits or regulates the flow of electric current in a circuit and remembers the amount of charge that has previously passed through it. In other words, they are non-volatile and that is very important because they retain memory without power.
It has been discovered that networks of oscillators controlled by memristors can come close to emulating the neural networks of the human brain which, it seems, actually operate on oscillatory signals. It’s an appealing notion, and work on artificial neutrons and synapses has been going on, with considerable success, for some years now. Joint Swedish/Japanese Topspin research shows that oscillators and oscillating circuits can perform complex calculations in a way that mimics the way human nerve and memory cells appear to do the same.
Quoted in the academic journal ‘Nature Materials’, Johan Åkerman, Professor of Applied Spintronics in the Department of Physics at the University of Gothenburg, says: “Computers are now incredibly good at performing advanced cognitive tasks, such as language recognition and images or the display of superhuman chess skills, thanks in large part to artificial intelligence.At the same time, the human brain is still unmatched in its ability to perform tasks efficiently and efficiently. Finding new ways to perform computations that mimic the energy-efficient processes of the brain has been a major research goal for decades.
He adds: “Cognitive tasks…require significant computing power, and mobile applications, in particular, such as drones and satellites, require energy-efficient solutions. This is an important advance because we show that it is possible to combine a memory function with a calculation function in the same component. These components function more like the energy-efficient neural networks of the brain, allowing them to become important building blocks in future more brain-like computers.
When it comes to better and more energy-efficient mobile handsets, Professor Åkerman believes the new research will lead to new features on the devices. He uses the example of digital assistants such as Siri where currently all processing has to be done by remote servers because such heavy processing work on a mobile handset is very energy inefficient. However, if the components could be small enough, several hundred of them would fit into a mobile handset enabling power-efficient local processing and obviating the need for power-hungry servers. Under laboratory conditions, the researchers were able to produce components so small under the microscope that they are “less than the size of a single bacterium”.
By the way, spintronics, or spin electronics, is a technology that exploits the intrinsic spin properties of the electron. An electron can exist in one of two spin states: spin-up and spin-down i.e. it can spin clockwise or counterclockwise of a watch with a constant frequency around its axis. This property can be used to represent a 0 or a 1 in logical operations.