Federal Grant Award Summary

The National Science Foundation's Division of Computing and Communication Foundations (CFDA 47.070 – Computer and Information Science and Engineering) awarded Clemson University a $368,275 Project Grant effective July 1, 2026, through June 30, 2031, to develop an automated, robust, and adaptive feature engineering framework for machine learning applications. The research will produce algorithmic prototypes and technologies that unify feature generation, feature transformation, and feature selection through a multi-agent reinforcement learning (MARL) framework. The core deliverables include novel reinforcement learning policies for operator agents, optimized action spaces through compression strategies, multi-level information sharing and interaction mechanisms for agent coordination, and high-efficiency computational improvements via cache and incremental update mechanisms. These technology outcomes are designed for adoption by commercial entities to enhance real-world products and services across critical application domains including disease diagnosis, vehicular security, emergency medical services, software engineering, material reliability, and electronic commerce.

In addition to research outputs, the award supports educational and workforce development objectives, including curriculum development, undergraduate student research opportunities, internships, and K-12 outreach initiatives. The award reflects the NSF's emphasis on advancing computational methodologies with practical impact while building a skilled workforce in computer and information science through mentoring and educational partnerships.