Funded by the U.S. Department of Energy (DOE), the project seeks to overcome the limitations of current neuromorphic models, which often rely on overly simplistic designs, by developing a fully analog neuromorphic chip. This chip will incorporate novel active dendritic processing and non-linear synaptic devices for incremental learning, allowing for more efficient and complex spatiotemporal information processing.
The ultimate goal is to improve the size, weight, and power (SWaP) efficiency of AI systems by leveraging the sophisticated computational mechanisms of biological intelligence.