This National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) Federal Grant Award under CFDA Program 47.070 provides $600,000 in funding to Carnegie Mellon University from June 1, 2025 to May 31, 2028 for a project titled "SMALL: VISUAL CORTICAL RECURRENT CIRCUITS FOR MANIFOLD LEARNING AND MEMORY ATTENTION." The project aims to investigate the computational mechanisms underlying the brain's ability to rapidly form local recurrent circuits in the early visual cortex, which enables the encoding of relationships between concepts. The research will develop a machine learning framework to formalize this rapid recurrent plasticity, testing the hypothesis that these recurrent circuits perform manifold transformations to bring semantically related concepts closer together in the neural activity latent space. The research will combine computational modeling with neurophysiological experiments, evaluating the practical utility of these biologically-grounded models in real-world computer vision applications. This work seeks to offer a new conceptual framework for understanding cortical recurrent circuits and inform the development of more efficient, generalizable, and brain-inspired artificial intelligence systems. No sub-awards are planned under this grant.
Mod # | Description | Reason For Modification | Federal Obligation (Click to sort descending) | Date (Click to sort ascending) |
|---|---|---|---|---|
| Not listed | $600.0k | 6/25/25 |