Neuromorphic computing is a method of computer engineering in which elements of a computer are modeled after systems in the human brain and nervous system. It aims to improve energy efficiency and computational power for AI tasks by mimicking biological neural architectures, using electronic circuits to simulate neurons and synapses.
The concept was pioneered by Carver Mead in the late 1980s. It represents a departure from the traditional von Neumann architecture, focusing on parallel processing and event-driven computation.
Neuromorphic chips like Intel's Loihi and IBM's TrueNorth are being used for edge AI applications, robotics, and sensory processing tasks where low power consumption and real-time processing are critical.