Welcome to this webpage! Thanks for your interest in our research!
My name is Wenzhu Bi. I has been a research assistant in the Computational Mathematics Master's Program at Duquesne University from 2002 to 2004. My advisor is Dr. Jeffrey Jackson. Our research interest is primarily in Disjunctive Normal Form(DNF) Learning in the field Computational Learning Theory.
My work is to program for some new algorithms proposed by Dr. Jackson, analyze the results and help to improve the algorithm. Our algorithm attempts to learn a monotone DNF f using a hypothesis h (h is not necessarily monotone DNF) in polynomial time from a set of uniformly-distributed samples generated by f. In our particular case, h is a threshold function, which is a sign of a sum of parity functions.
Please notice that in our tests, we perform this exact computation for the Fourier coefficients from the whole truth table so that we can leave out the variability added by estimating the Fourier coefficients. In an actual implementation, these coefficients would be estimated using polynomial-size set of examples (sufficient for PAC learning by Chernoff bound).
My Thesis Defense Slides give an introduction of the problem, the proposed algorithm and the results; You can also refer to my Thesis to get more detail information and see the Original Results Files for your own analysis. You can also download the source code: Linux-compressed version and Windows-compressed version( the file "readme.txt" explains how to run the program).Please feel free to contact me if you have any questions. My email is wenzhubi@yahoo.com.
Thanks again for visiting the webpage!