Bio-circuitry simulates synapses, neurons and a first step towards sensory computing
Researchers from the Department of Energy’s Oak Ridge National Laboratory and Texas A&M University have demonstrated bio-inspired devices which accelerate the route to neuromorphic computing, or brain-like computing.
Nature Communications published the first results of a \”memcapacitor\” – a charge storage device with memory, which processes information like synapses in the brain. Their discovery could help to support the development of computing networks that are modeled after biology, for a sensor-based approach to machine learning.
The goal of the research is to create materials and computing elements with a large degree of interconnectedness and flexibility that mimic biological synapses, neurons, and other neural structures. This will enable autonomous systems which operate differently from current computing devices, offering new functionality and learning abilities, said Joseph Najem. He was a postdoctoral fellow at ORNL’s Center for Nanophase Materials Sciences (a DOE Office of Science user facility) and currently an assistant professor of mechanical Engineering at Penn State.