CAREER: CORTEX: Next-Generation COmputing with Brain-Inspired Resilient, Transformative, and EXplainable Intelligence

This project introduces a new computational framework towards building resilient, energy-efficient, and neuroscience-aligned artificial intelligence (AI) systems by drawing inspiration from biological nervous systems. Modern AI systems are powerful, but they depend on large datasets, extensive computing resources, and high energy consumption. Furthermore, current AI systems struggle to adapt to changing environments and new tasks. This project translates brain-inspired principles that underlie biological intelligence into new AI algorithms that can sense, remember, learn, and adapt with improved efficiency and reliability compared with current data- and energy-intensive approaches. Educational activities in this project introduce students to the connections among AI, neuroscience, and computing while creating research experiences across K–12, undergraduate, and graduate education. The project develops a unified brain-inspired computing framework organized around three principles that are a core part of biological nervous systems. In Aim 1, this project develops hippocampus-inspired learning algorithms that capture temporal relationships among events, spatial context, and prior experience, enabling real-time learning and adaptability. In Aim 2, this project develops learning algorithms inspired by the functional diversity of biological neuron types to improve energy efficiency, robustness, and generalizability for efficient learning and performance. In Aim 3, this project designs retina-inspired computing modules and lightweight learning architectures to support fast visual perception by converting visual input into sparse, task-relevant feature events, similar to early processing in the biological retina. Together, the project provides computational testbeds to examine hypotheses about cortical computation, sensory encoding, memory formation, and adaptive learning, while also informing educational activities that connect brain-inspired AI, neuroscience, and learning science. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria. NSF Award ID: 2539714 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Maryam Parsa | Institution: George Mason University, FAIRFAX, VA | Award Amount: $655,286 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2539714 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2539714.html