We all know that besides industrial applications, burning laser pointer technology has been widely used in many fields such as research, biomedicine and communications, which has played a great role in promoting the development of human science and technology.

The research team from Princeton University in the United States has developed the world's first silicon photonic neuromorphic chip and has proven that it can speed up the operation nearly 2000 times, helping to promote artificial intelligence such as facial recognition, object recognition, natural language processing, machine translation Technical application development.

The research team etched each node of the new chip in the form of a mini circular waveguide into a silicon base, allowing light to circulate through it. When light is input to the node, the output of the laser working at the node threshold is modulated, and the output of the high power laser is fed back to the node, creating a feedback circuit with non-linear features. With regard to the extent to which this nonlinear behavior mimics neural behavior, researchers have demonstrated that its output is mathematically equivalent to a "continuous-time recurrent neural network." Researchers used a network of 49 photon nodes to simulate the neural network and used it to solve the mathematical problems of differential equations. They found that this silicon photon neuromorphic chip can compare the arithmetic speed up 1960 times.