Research

Current Research Activities of
Mathukumalli Vidyasagar
Fellow of The Royal Society
Cecil & Ida Green Chair in
Systems Biology Science


Notes

I specifically chose the above photo, crow's feet and all, just to highlight the point that research should be fun!

This page contains a description of my current research interests, recent papers and papers awaiting publication, and recent seminar presentations. Reprints of scientific papers that are already published can be found under "publications" while other writings can be found under (what else?) "other writings."

Contents

Current research interests
Recent Papers (updated frequently)
Slides of some recent talks

Current Research Interests

My research interests are in the broad area of system and control theory, and its applications. After spending an entire lifetime "ducking" probability theory, at the moment I am working on problems in stochastic realization, specifically approximating high-order Markov and hidden Markov processes by lower-order processes, and the large deviation properties of such processes. Recently I have also developed an interest in compressed sensing theory. On the applications front, I am interested in applying these (or any other appropriate) ideas to problems in computational biology with emphasis on cancer.

Recent Papers

Preprints

  • M. Eren Ahsen and M. Vidyasagar, "Near-ideal behavior of compressed sensing algorithms." Preprint at arxiv, 1401.6623. PDF
  • M. Vidyasagar, "Machine learning methods in the computational biology of cancer," Preprint at arxiv, 1402.5782. PDF

Journal Papers

  • M. Vidyasagar, "An elementary derivation of the large deviation rate function for finite state Markov processes," Asian Journal of Control, 16(1), 1-19, January 2014. PDF
  • M. Eren Ahsen and M. Vidyasagar, "Mixing coefficients between discrete and real random variables: Computation and properties", IEEE Transactions on Automatic Control, 59(1), 34-47, January 2014. PDF
  • M. Vidyasagar, "A metric between probability distributions on finite sets of different cardinalities and applications to order reduction", IEEE Transactions on Automatic Control, 57(10), 2464-2477, October 2012. PDF
    For a longer version with more details, see the arXiv version
  • M. Vidyasagar, "Probabilistic methods in cancer biology", European Journal of Control, 17(5-6), 483-511, September - December 2011. PDF
  • M. Vidyasagar, "The complete realization problem for hidden Markov models: A survey and some new results", Mathematics of Control, Signals and Systems, 23(1), 1-65, 2011. PDF

Conference Papers

  • M. Eren Ahsen, Nitin K. Singh, Todd Boren, M. Vidyasagar and Michael A. White, "A new feature selection algorithm for two-class classification problems and their application to endometrial cancer", IEEE Conference on Decision and Control, Maui, Hawaii, December 2012, PDF
  • M. Vidyasagar and Yutaka Yamamoto, "Convergence and compactness of families of proper plants in the graph topology", IEEE Conference on Decision and Control, Maui, Hawaii, December 2012, PDF
  • M. Vidyasagar, "A metric between probability distributions on IEEE Conference on Decision and Control, and European Control Conference, Orlando, FL, 710-715, December 2011. PDF
  • M. Vidyasagar, "Optimal order reduction of probability distributions by maximizing mutual information", IEEE Conference on Decision and Control, and European Control Conference, Orlando, FL, 716-721, December 2011. PDF
  • Yutaka Yamamoto and M. Vidyasagar, "Compact sets in the graph topology and applications to approximation of system design", IEEE Conference on Decision and Control, and European Control Conference, Orlando, FL, 621-626, December 2011. PDF
  • Kun Deng, Prashant G. Mehta, Sean P. Meyn and M. Vidyasagar, "A recursive learning algorithm for model reduction of hidden Markov models", IEEE Conference on Decision and Control, and European Control Conference, Orlando, FL, 4674-4679, December 2011 PDF

Slides of Some Recent Talks

  • "Recent Developments in Compressed Sensing," Purdue University, 11 October 2013. PDF
  • "Machine Learning Methods in Cancer Biology," Rufus Oldenburger Memorial Lecture, Purdue University, 10 October 2013. PDF
  • "Predicting Extreme Events in Finance and Weather: Use of Heavy-Tailed Distributions," University of Buffalo, 03 October 2013. PDF
  • "How Engineers Can Assist in Finding the Right Therapies for Cancer", Rufus Oldenburger Lecture, American Society of Mechanical Engineers, Ft. Lauderdale, FL, 18 October 2012. PDF
  • "Probabilistic Methods in Cancer Biology", Center for Control Systems and Technology Seminar, UT Dallas, 08 October 2012. Note: This talk is completely different from my Plenary Talk at CDC 2011, though I used the same title. PDF
  • "Predicting the Risk of Lymph Node Metastasis in Endometrial Cancer Using Machine Learning Techniques", New Fellows Presentation at The Royal Society, 12 July 2012. PDF The audio of the talk can be found here
  • "Probabilistic Methods in Cancer Biology", Plenary Talk at the 2011 IEEE Conference on Decision and Control and 20th European Control Conference, Orlando, FL, 14 December 2011. PDF
  • "Four Decades of Control: A Journey of Reinventions", Harry H. Nyquist Lecture, Division of Dynamical Systems, Measurement and Control, American Society of Mechanical Engineers, Washington, DC, 01 November 2011. PDF
  • "Metric Distances Between Probability Distributions of Different Sizes", talk given at Johns Hopkins University on 20th October 2011. PDF
  • "Predicting Extreme Events in Finance, Internet Traffic and Weather", IEEE Control Systems Society Webinar, 12 November 2010. PDF Webinar on IEEE CSS On-Line Lecture Library here