SB AdvisoryMay 20, 2020 update: Keep up with the latest from Stony Brook about the coronavirus situation.  More information
Skip Navigation
Search
Peter Beling, PhD
Distringuished Professor
University of Virigina (UVA)

Peter A. Beling is a professor in the Department of Engineering Systems and Environment at the University of Virginia (UVA). Dr. Beling’s research interests are in the area of decision-making in complex systems, with emphasis on adaptive decision support systems and on model-based approaches to system-of-systems design and assessment. His research has found application in a variety of domains, including mission-focused cybersecurity, reconnaissance and surveillance, prognostic and diagnostic systems, and financial decision making. He directs the UVA site of the Center for Visual and Decision Informatics, a National Science Foundation Industry/University Cooperative Research Center, and the Adaptive Decision Systems Laboratory, which focuses on data analytics and decision support in cyber-physical systems. Dr. Beling has served as editor and reviewer for many academic journals and as a member of several National Academies panels. Dr. Beling received his Ph.D.in operations research from the University of California at Berkeley.

Contact Information: pb3a@virginia.edu

Abstract

A Systems Theoretic Perspective on Transfer Learning

The machine learning formulation of transfer learning is incomplete from a systems theoretic perspective. It focuses on algorithm parameters, features, and samples, and neglects the perspective offered by considering system structure and system dynamics. Furthermore, while the machine learning formulation serves to classify methods and literature, the systems theoretic formulation we propose serves to provide a framework for the top-down design of transfer learning systems, including a novel definition of transfer learning and identification of key design parameters. We dichotomize the transfer learning problem into a question of transferring system structure and dynamics. We formulate our framework in the context of input-output systems and discuss results for several real-world systems.

View Presentation