Bio

​B.A. University of Virginia, 1985M.S. ​George Washington University, 1987Ph.D. ​University of California at Berkeley, 1992

"My research and teaching centers on machine learning and optimization methods to support decision-making in complex systems."

Peter A. Beling, Professor

Peter A. Beling is Professor and Associate Chair for Research in the Engineering Systems and Environment Department. Dr. Beling’s research interests are in the area of decision-making in complex systems, with emphasis on machine learning and 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 is the co-founder of the Financial Decision Engineering research group at UVa, which is a focal point for research on the mathematical modeling and risk management aspects of consumer and retail credit. Dr. Beling has served as editor and reviewer for many academic journals and has served as a member of five National Research Council panels. He received the Ph.D. in Operations Research from the University of California at Berkeley, M.S. in Operations Research from The George Washington University, and B.S. in Mathematics from the University of Virginia.

Research Interests

  • Cybersecurity
  • Autonomy and Controls/Control Systems
  • Human Machine Interface
  • Computational Statistics and Simulation/Statistical Modeling
  • Machine Learning

Selected Publications

  • The WEAR Methodology for PHM Implementation. Journal of Manufacturing Systems, to appear. 2017. Adams, S., Malinowski, M., Heddy, G., Choo, B., and Beling, P.
  • Multi-agent Inverse Reinforcement Learning for Zero-sum Games. IEEE Transactions on Computational Intelligence and AI in Games. (published online) 2017. Lin, X., Beling, P., and Cogill, R.
  • Optimal Model-Based (3D/6D) Pose Estimation with Structured-Light Depth Sensors. IEEE Transactions on Computational Imaging, pp. 1-14 (published online). 2017. Landau, M. and Beling, P.
  • Adaptive Multi-scale PHM for Smart Manufacturing Systems. International Journal of Prognostics and Health Management, pp. 1-15 (published online). 2016. Choo, B., Adams, S., Marvel, J., Weiss, B., and Beling, P.
  • Feature Selection for Hidden Markov Models and Hidden Semi-Markov Models. IEEE Access, 4, pp. 1642-1657. 2016. Adams, S., Beling, P., and Cogill, R.
  • Regulatory Capital Decisions in the Context of Consumer Loan Portfolios. Journal of the Operational Research Society, pp. 1-20 (published online). 2016. Rajaratnam, K., Beling, P., and Overstreet, G.
  • Adaptive Multi-scale Optimization: Concept and Case Study on Simulated UAV Surveillance Operations. IEEE Systems Journal, pp. 1-12 (published online). 2015. Reyes, I., Beling, P., and Horowitz, B.
  • Task Recognition from Joint Tracking Data in an Operational Manufacturing Cell. Journal of Intelligent Manufacturing, pp. 1-15 (published online). 2015. Rude, D., Adams, S., and Beling, P.
  • Gaussian Process-Based Algorithmic Trading Strategy Identification. Quantitative Finance, 15, no. 10, pp. 1683-1703. 2015. Yang, S., Qiao, Q., Beling, P., Scherer, W., and Kirilenko, A.
  • Recognition of Agents from Observation of Their Sequential Behavior. European Conference on Machine Learning and Principles and Practices of Knowledge Discovery in Databases (ECML-PKDD 2013), pp. 33-48. 2013. Qiao, Q. and Beling, P.
  • Inverse Reinforcement Learning with Gaussian Processes. American Control Conference (ACC 2011), pp. 113-118. 2011. Qiao, Q. and Beling, P.
  • Decentralized Bayesian Search using Approximate Dynamic Programming Methods. IEEE Transactions on Systems, Man and Cybernetics, Part B, 38 (4), pp. 970-975. 2008. Zhao, Y., Patek, S., and Beling, P.
Google Scholar
ResearchGate

Courses Taught

  • SYS 3021 – Deterministic Decision Models Fall semester
  • SYS 6054 – Financial Engineering
  • SYS 6581/7581 – Network/Combinatorial Optimization
  • SYS 6581 – Production Systems

Featured Grants & Projects

  • Prognostics and Health Management (PHM)

    This research centers on the development of models and algorithms to support the next generation of prognostics and health management (PHM) decision systems. Our approaches are primarily data-driven and make use of reinforcement learning and other machine learning methods for prediction and control. Application domains include: smart manufacturing, funded by NIST; machining, funded by the the Commonwealth Center for Advanced Manufacturing (CCAM); marine actuators, funded by Luna Innovations; and energy harvesting structures, funded by Luna Innovations.

  • Adaptive Improvement in Manual Production Processes

    The goal of this research is to develop repeatable methods for identifying training and process interventions that are likely to reduce the variability of manual manufacturing processes. My group has developed an inverse reinforcement learning approach in which automated inference about worker activity – based on computer vision and activity recognition algorithms we have developed – is combined with measurements from correlated process events to make inferences about the cognitive goals of workers. The hypothesis is that detailed understanding of worker cognitive goals can be the basis for training and process interventions that improve quality and productivity. This work is funded by Rolls Royce Corporation under the University Technology Center.

  • Cyber Security of Cyber Physical Systems

    The past few years have seen a dramatic rise in interest in providing cyber security protection for cyber physical systems, such as cars, drones, and power stations. Other researchers at UVA have pioneered theory and methods for defending specific functions of the physical system rather than the entire computer systems and networks that support all functions. Given a set of risks, a library of potential defensive measures, and a finite security budget, the question becomes which of the functions to protect. My group has developed new decision-making methodologies to address this question. This work is funded by the DoD through the Systems Engineering Research Center.