Researchers claim analog resistor network enables processor-less machine learning. #AIrevolution

Researchers have developed an analog network of resistors that could potentially perform machine learning tasks without the need for a traditional processor. This new approach to computing, known as “resistive processing units,” relies on the interactions between resistors to perform computations. By leveraging the physical properties of resistors, this analog network can efficiently process data and perform tasks typically associated with machine learning algorithms. The researchers behind this technology believe that it could lead to more energy-efficient and faster computing systems. This innovation represents a departure from the digital computing paradigm that has dominated the industry for decades. Instead of relying on binary calculations, this analog network operates on continuous values, mimicking the way the human brain processes information. While this technology is still in the early stages of development, it holds promise for revolutionizing the field of artificial intelligence and machine learning. By harnessing the power of resistors in a networked system, researchers hope to unlock new possibilities for computing that were previously thought impossible.

Source link

Source link: https://www.hackster.io/news/an-analog-network-of-resistors-promises-machine-learning-without-a-processor-researchers-say-d9cb0655b7a0R


https://www.hackster.io/news/an-analog-network-of-resistors-promises-machine-learning-without-a-processor-researchers-say-d9cb0655b7a0.amp