Contact Information

Department of Computer Science and Engineering • Lopata Hall, Room 526
Washington University in St. Louis • One Brookings Drive • St. Louis, MO 63130

Research Interests

Any real-time system, such as a mobile robot, packet routing network, or traffic controller, must be able to ensure timely execution despite uncertainty in the behavior of its components and variability in its operating environment. Reinforcement learning combines tools from machine learning, operations research, and optimal control to optimize the behavior of controllers in the face of uncertainty. My research interests lie in using techniques from reinforcement learning to address issues that arise in the design of real-time systems, particularly resource allocation among stochastic tasks in variable environments, and adapting to non-stationarity in the task behavior. In my dissertation, I addressed the problem of weighted fair-share scheduling of a shared resource among several non-preemptible tasks with stochastic execution requirements. This involved the automatic discovery of appropriate problem representations as well as learning task behavior from observations of the controlled system. This research represents a jumping-off point to address a wider range of concerns that arise in real-time systems

Education

2002 - 2009 • Doctor of Philosophy, Computer Science • Washington University in St. Louis
1997 - 2001 • Bachelor of Science, Computer Science • University of Nebraska - Lincoln

Teaching Experience

• Summer 2006 • Fundamentals of Computer Science

  The purpose of this class is to introduce advanced students with a bachelor's degree or similar in an unrelated field to the central concepts of computer science, such as discrete mathematics, propositional logic, and the design and analysis of algorithms and data structures.

• Spring 2006 • Seminar on Machine Learning

  This seminar was organized to survey advanced topics in reinforcement learning.

Publications

Scheduling Policy Design using Stochastic Dynamic Programming, Robert Glaubius. Ph.D. Thesis, Washington University in St. Louis, St. Louis, MO, USA, 2009. [download pdf]

"Scheduling Policy Design for Autonomic Systems", Robert Glaubius, Terry Tidwell, Christopher Gill, and William D. Smart. International Journal on Autonomous and Adaptive Communications Systems, 2(3):276-296, 2009. [download pdf]

"Scheduling Design with Unknown Execution Time Distributions or Modes", Robert Glaubius, Terry Tidwell, Christopher Gill, and William D. Smart. Technical Report WUCSE-2009-15, Department of Computer Science and Engineering, Washington University in St. Louis, 2009. [download pdf]

"Scheduling Design and Verification for Open Soft Real-time Systems". Robert Glaubius, Terry Tidwell, William D. Smart, and Christopher Gill. In "Proceedings of the 2008 Real-Time Systems Symposium (RTSS 2008)", pages 505-514, 2008. [download pdf]

"Scheduling as a Learned Art", Christopher Gill, William D. Smart, Terry Tidwell, and Robert Glaubius. In "Proceedings of the 4th International Workshop on Operating Systems Platforms for Embedded Real-Time Applications (OSPERT 2008)", pages 53-62, 2008. [download pdf]

"Scheduling for Reliable Execution in Autonomic Systems". Terry Tidwell, Robert Glaubius, Christopher Gill, and William D. Smart. In "Proceedings of the 5th International Conference on Autonomic and Trusted Computing (ATC-08)", volume 5060, pages 149-161, 2008. [download pdf]

"Manifold Representations for Value Function Approximation in Reinforcement Learning", Robert Glaubius and William D. Smart. Technical Report WUCSE-2005-19, Department of Computer Science and Engineering, Washington University in St. Louis, 2005. [download pdf]

"Speeding up Reinforcement Learning using Manifold Representations: Preliminary Results", Robert Glaubius, Motoi Namihira, and William D. Smart. International Joint Conference on Artificial Intelligence (IJCAI 2005) Workshop on Reasoning with Uncertainty in Robotics, 2005. [download pdf]

"Manifold Representations for Value Function Approximation", Robert Glaubius and William D. Smart. In Proceedings of the the AAAI-04 Workshop on Learning and Planning in Markov Processes, pages 13-18, 2004. [download pdf]

"Constraint Modeling and Reformulation in the Context of Academic Task Assignment", Robert Glaubius and Berthe Y. Choueiry. ECAI 2002 Workshop on Modelling and Solving Problems with Constraints, 2002.