Researchers at the Chinese University of Hong Kong have developed a laser-based artificial neuron that mimics the functions of a biological nerve cell, with a processing speed of 10 GBaud, a billion times faster than its natural counterpart. Led by Chaoran Huang, the team created a chip-based quantum-dot laser graded neuron that can process vast amounts of data quickly and efficiently. This technology has the potential to advance artificial intelligence tasks such as pattern recognition and sequence prediction. The laser graded neuron overcomes the speed limitations of current photonic versions of spiking neurons, making it ideal for high-speed reservoir computing. Huang believes that integrating this technology into edge computing devices could facilitate faster and smarter AI systems with reduced energy consumption. The research was published in Optica, a journal of the Optica Publishing Group, and demonstrates exceptional performance in AI tasks such as arrhythmia detection and image classification.
↧