Isbell’s research explores applications of statistical machine learning to human-computer and human-human interaction. It revolves around a vision for building adaptive and personalized environments that model humans and then uses those models to improve their own behavior. “The core technical nuggets,” says Isbell, “are algorithms for activity discovery, as opposed to activity recognition, as well as the integration of such algorithms into an actual programming language.”

Agents that can build their own representations of human behavior have great potential to impact entertainment, education and life management. The programming language developed in part under Isbell’s grant will also provide a mechanism for programmers with little expertise in machine learning to develop such agents more easily, and may offer opportunities for new ways of teaching programming.

NSF CAREER Awards are one of the highest honors given to young university faculty in science and engineering. For more information about Charles Isbell, click here.