Welcome to BrAIN Web
The human brain has immense learning capability at extreme energy efficiency and scale, which no artificial system has been able to match. Although reverse-engineering the brain has long been one of the top priorities of science and technology, the conventional methods based on electronics failed to match the scalability, energy efficiency, and self-learning capabilities of the human brain. Moreover, the human brain is remarkably fast at learning in a manner that flexibly adapts to new situations and tasks, but current approaches in Artificial Intelligence (AI) often require extensive training and can be brittle in the face of unexpected changes requiring flexibility.
The Center for Brain, AI, and Neuromorphic Computing (BrAIN) research will pursue a brain-derived—rather than a brain-inspired—architecture enabling the development of intelligent agents that can learn and generalize as humans in unpredictable environments. At the forefront of this research endeavor is the development of new photonic and electronic memristive materials, device technologies, and 3D electronic-photonic integrated circuits (3D EPICs). These technologies will enable novel brain-derived neuromorphic computing hardware with energy efficiency, connectivity, density, and scalability comparable to the human brain. Their combination with bio-realistic learning algorithms and architectures may yield an “artificial brain” with general self-learning capabilities that can then be used to test causal hypotheses from neuroscience.