Probabilistic Programming for Advancing Machine Learning (PPAML)*
Abstract
*Previously funded in PE 0602702E, Project TT-13. The Probabilistic Programming for Advancing Machine Learning (PPAML) program will create an advanced computer programming capability that greatly facilitates the construction of new machine learning applications in a wide range of domains. This capability will increase the number of people who can effectively contribute, will make experts more productive, and will enable the creation of new tactical applications that are inconceivable given today's tools. The key enabling technology is a new programming paradigm called probabilistic programming that facilitates the management of uncertain information. In this approach, developers will use the power of a modern (probabilistic) programming language to quickly build a generative model of the phenomenon of interest as well as queries of interest, which a compiler will convert into an efficient application. PPAML technologies will be designed for application to a wide range of military domains including ISR exploitation, robotic and autonomous system navigation and control, weather prediction, and medical diagnostics.
Document Details
- Document Type
- Accomplishment
- Publication Date
- Oct 01, 2015
- Source ID
- 8f78ddda69fb6a8fc0bae5b33c8ae0eb