[Ml-stat-talks] Statistics Seminar at ORFE

Jelena Bradic jbradic at Princeton.EDU
Fri Oct 1 11:31:33 EDT 2010


Dear all,

We will be having a Statistics Seminar today, at 12:30pm, in  
Statistics Lab, Sherrerd Hall 213.
The speaker is Yichao Wu from North Carolina State University.



Title: Robust Model-Free Multiclass Probability Estimation


Abstract: Classical statistical approaches for multiclass probability  
estimation are typically based on regression techniques such as  
multiple logistic regression, or density estimation approaches such  
as linear discriminant analysis (LDA)  and quadratic discriminant  
analysis (QDA). These methods often make certain assumptions on the  
form of probability functions or on the underlying distributions of  
subclasses. In this article, we develop a model-free procedure to  
estimate multiclass
probabilities based on large-margin classifiers. In particular, the  
new estimation scheme is employed by solving a series of weighted  
large-margin classifiers and then systematically extracting the  
probability information
from these multiple classification rules. A main advantage of the  
proposed probability estimation technique is that it does not impose  
any strong parametric assumption on the underlying distribution and  
can be applied for a wide  range of large-margin classification  
methods. A general computational algorithm is developed for class  
probability estimation. Furthermore, we establish asymptotic   
consistency of the probability estimates. Both simulated and real  
data examples are presented
to illustrate competitive performance of the new approach and
compare it with several other existing methods.

This is a joint work with Hao Helen Zhang and Yufeng Liu



--------------------------------------
Jelena Bradic
Department of Operations Research and Financial Engineering
Princeton University, Princeton, NJ 08544
jbradic at princeton.edu



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cs.princeton.edu/pipermail/ml-stat-talks/attachments/20101001/843e4c9e/attachment.html>


More information about the Ml-stat-talks mailing list