Professor John Oommen gave a distinguished research seminar in CIS

February 21, 2020

Dr. John Oommen, Chancellor's Professor of Carleton University, Life Fellow of IEEE and Fellow of LAPR, visited CIS and gave a distinguished research seminar on anti-Bayesian statistical pattern recognition.

Dr. John Oommen, Chancellor's Professor of Carleton University, Life Fellow of IEEE and Fellow of LAPR, visited CIS and gave a distinguished research seminar on anti-Bayesian statistical pattern recognition. 

The gold standard for a classifier is the condition of optimality attained by the Bayesian classifier. In the presentation, Dr. Oommen showed that we can obtain optimal results by operating in a diametrically opposite way, i.e., a so-called “anti-Bayesian” manner. Indeed, he showed the completely counter-intuitive result that by working with a very few (sometimes as small as two) points distant from the mean, one can obtain remarkable classification accuracies.  Further, if these points are determined by the Quantiles of the distributions, the accuracy of the new method, referred to as Classification by Quantile Statistics (CQS), attains the optimal Bayes' bound. This claim has been proven for many uni-dimensional, and some multi-dimensional distributions within the exponential family, and the theoretical results have been verified by rigorous experimental testing. They have also been tested on real-life data sets.

Dr. Oommen met CIS faculty members and students during his visit. Several potential future collaborations have been identified.

Back to top of page