search menu icon-carat-right cmu-wordmark

The Promise and Challenges of Recommendation Systems for the DoD

Presentation
This presentation describes recommendation systems, their potential use by the Department of Defense (DoD), and the impact of CMU SEI's research in this area.
Publisher

Software Engineering Institute

Watch

Abstract

This presentation explores concepts in recommendation systems, with a focus on data motion (the most expensive part of a calculation for a recommendation system). The topics addressed include the concepts and foundations of recommendation systems, their applications to Department of Defense (DoD) and the intelligence community, Facebook's Deep Learning Recommendation Model (DLRM), and MLPerf. Advanced computing in the SEI Emerging Technology Center (ETC), CMU SEI advances in DLRM, and the impact of this research work round out the presentation.