Chinese researchers have developed a laser-based artificial neuron that mimics the functions of biological neurons and processes data one billion times faster than natural counterparts. This breakthrough could revolutionize artificial intelligence and advanced computing technologies. Chaoran Huang, leader of the research team at the Chinese University of Hong Kong, stated, "Our laser neuron has the potential to work faster by overcoming the speed limitations of existing photonic neurons."
The technology aims to make artificial intelligence decision-making processes both faster and more efficient.
Biological neurons are divided into two main types: graded and spiking. Graded neurons process signals in a precise and continuously changing manner, while spiking neurons either fully transmit or do not transmit signals. The newly developed laser-based neuron simulates the operation of graded neurons, offering superior speed and accuracy.
Developed with chip-based quantum laser technology, the laser neuron can process signals at a rate of 10 gigabits per second. This enables it to process 100 million heartbeats or 34.7 million digital handwritten images in just one second.
Researchers believe this speed will enable more efficient use in artificial intelligence applications.
Laser-based neurons offer fast and efficient computing while reducing energy consumption. The research team created a reservoir computing system with their laser neurons, which detected heart arrhythmias with 98.4% accuracy and excelled in tasks like image classification due to its high-speed data processing capability.
Chaoran Huang said, "A single laser-based neuron can behave like a small neural network. By combining multiple laser neurons, we can create a network similar to the one in the biological brain, composed of billions of neurons, in the field of artificial intelligence."
The researchers aim to further increase the speed of laser-based neurons and develop deep reservoir computing architectures by combining multiple laser neurons. This technology is expected to revolutionize artificial intelligence applications, especially those requiring real-time decision-making and energy efficiency.
