DATA-DRIVEN ALGORITHM DESIGN: MATHEMATICAL AND COMPUTATIONAL FOUNDATIONS
Abstract
Obtaining expertise relies on a detailed understanding of the structure underlying the problem in the context of a sensing-to-learning-to-actuation loop. A new approach to data-driven design of learning and inference algorithms is required in which procedures are learned directly from data, operate within a specified computational budget, and come with statistical performance guarantees. Motivated by these considerations the proposed research agenda is to develop new computational foundations for learning new algorithms directly from data.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 12, 2021
- Source ID
- FA95502010320
Entities
People
- Venkat Chandrasekaran
Organizations
- Air Force Office of Scientific Research
- California Institute of Technology
- United States Air Force