Researchers at the University of Virginia’s School of Education and Human Development and School of Engineering and Applied Sciences have been awarded a new $1.4 million grant from the Bill & Melinda Gates Foundation to use artificial intelligence to improve teachers’ math instruction. We plan to improve the quality.
The Artificial Intelligence to Advance Large-Scale Instruction project develops an AI model that reviews videos of math teachers in their classrooms and evaluates specific elements of instruction, from whole-group instruction to individual student work. and test. The team will also develop a dashboard that will provide teachers with relevant feedback in real time to improve their instruction.
“Classroom observation videos are an incredibly useful tool for improving instructional skills, but evaluating them is an incredibly time-consuming process,” said Dr. Peter Youngs said. “By leveraging AI to classify videos, we hope to provide teachers with meaningful feedback that is not only actionable, but also accessible and affordable.”
The grant is based on cutting-edge research by co-principal investigator Scott T. Acton, a professor in the School of Engineering. Acton first engineered AI breakthroughs in human motion detection technology, applying neural networks to video analysis with astonishing accuracy.
“This is about making professional development more accessible and equitable while supporting teachers to do what they do best: teach,” Acton said. “Combining AI technology with real-world educational challenges is a great way to impact classrooms across the country.”
By combining Youngs’ expertise with Acton’s engineering innovation, this project creates a seamless blend of educational insight and advanced technology. This collaboration, which also includes Assistant Professor Jonathan Foster from the University at Albany, ensures that this tool is practical and meaningful to teachers in diverse classrooms.
This AI tool uses synchronized video and audio to assess classroom dynamics, capturing details such as the quality of teacher-student interactions, lesson rigor, and instructional patterns. It provides richer and more comprehensive insights compared to traditional systems that rely solely on video and audio. When combined with coaching, the feedback from the tool is designed to improve the teaching skills and quality of elementary school mathematics teachers.
This program may be expanded to include English, history, and science applications in the future.
