Researchers from Northwestern University, Boston College, and MIT have developed a groundbreaking synaptic transistor inspired by the human brain, demonstrating advanced associative learning capabilities. Unlike previous brain-like computing devices, this transistor remains stable at room temperature, operates at high speeds, consumes minimal energy, and retains information without power. Published in Nature, the study challenges traditional computing architectures, emphasizing the transistor’s concurrent memory and information processing, mirroring the brain’s efficiency.
Driven by the need for energy-efficient alternatives in the age of big data, the researchers utilized moiré patterns, combining bilayer graphene and hexagonal boron nitride. Twisting these materials created unique electronic properties for neuromorphic functionality at room temperature. The transistor, trained to recognize patterns, exhibited superior associative learning even with incomplete input, showcasing potential applications in complex scenarios, such as self-driving vehicles navigating challenging conditions. This innovation marks a significant leap in AI technology towards higher-level thinking.