Mining and Understanding Software Enclaves (MUSE)
Abstract
The Mining and Understanding Software Enclaves (MUSE) program will develop program analyses and frameworks for improving the resilience and reliability of complex applications. MUSE techniques will apply machine learning algorithms to large software corpora to repair likely defects and vulnerabilities in existing programs and to discover new programs that conform to desired behaviors and specifications. MUSE frameworks will enable robust execution of large-scale and data-intensive computations. Specific technical challenges include persistent semantic artifact generation and analysis, defect identification and repair, pattern recognition, and specification inference and synthesis. MUSE research will improve the security of intelligence-related applications and enhance computational capabilities in areas such as graph processing, entity extraction, link analysis, high-dimensional data analysis, data/event correlation, and visualization. This program is an outgrowth of Probabilistic Programming for Advancing Machine Learning (PPAML).
Document Details
- Document Type
- Accomplishment
- Publication Date
- Oct 01, 2015
- Source ID
- e0ef4cb5563ebe817de741d83025542b