ECE 598 RT - Randomized Algorithms for System Design: Theory and Applications

Summer 2009 | Fall 2009 | Spring 2010 | Summer 2010
Section Type Times Days Location Instructor
RT LEC 1000 - 1050 M W F   143 Everitt Lab  Roberto Tempo
M Başar

Official Description Subject offerings of new and developing areas of knowledge in electrical and computer engineering intended to augment the existing curriculum. See Class Schedule or departmental course information for topics and prerequisites. May be repeated in the same or separate terms if topics vary.
Hours 0 to 4 hours.
Course Prerequisites Credit in ECE 313
Credit in ECE 515
Course Directors
Description This half-semester course will provide a broad perspective of the emerging area of randomization of uncertain systems with the objective to reduce the complexity of feedback, and discuss several topics, which include the connections between randomized algorithms and statistical learning theory, and the computation of the sample complexity. In particular, the course will address systems analysis and design using sequential and non-sequential randomized methods, and analyze advantages and disadvantages of both approaches.
Notes Course grade will be based on performance in homework assignments and the term project.
Credit 2 hours.
Topics
  • Uncertainty and Robustness
  • Probabilistic Robustness Analysis
  • Monte Carlo and Las Vegas Randomized Algorithms
  • Probability of Performance and Worst-Case Performance
  • Sample Complexity and Chernoff Bounds
  • Computational Complexity of Randomized Algorithms
  • Multivariate Random Vector and Matrix Generation
  • Quasi-Monte Carlo Methods
  • Probabilistic Robust Synthesis
  • Quadratic Performance and Convexity
  • Sequential Methods for Convex Problems (basic and advanced techniques)
  • Probabilistic LPV Systems
  • Non-sequential Methods for Convex Problems
  • Non-Sequential Methods for Nonconvex Problems
  • Optimization of Nonconvex Problems
  • Feasibility of Nonconvex Problems
  • Statistical Learning Theory
  • RACT: Randomized Algorithms Control Toolbob
  • Applications: aerospace and automotive control, multiagent systems, network control, quantized and switched systems, embedded and distributed systems
  • Robust and Randomized Algorithms for Control of Mini-UAVs (Unmanned Aerial Vehicles)
  • PageRank Computation in Google
Course Prerequisites A basic knowledge of probability theory (at the level of ECE 313) and familiarity with state space based control system analysis and design (at the level of ECE 515).
Texts There will be no required textbook for the course, but readings from the current literature will be assigned. Three key references, the materials from which will be used are given below (which will also be made available electronically).