Computer Science Colloquium
A Framework for Learning Predictive Structures from
Multiple Tasks and Unlabeled Data
IBM TJ Watson Research Center
Wednesday, March 7, 2005 11:15 A.M.
Room 1302 Warren Weaver Hall
251 Mercer Street
New York, NY 10012-1185
Colloquium Information: http://cs.nyu.edu/csweb/Calendar/colloquium/index.html
Richard Cole firstname.lastname@example.org, (212) 998-3119
Semisuperised learning, i.e. statistical machine
learning methods that use
both labeled and unlabeled data, have been actively
investigated in recent
years. However, at the current stage, we still don't
have a complete
understanding of their effectiveness.
This talk presents a closely related problem, which
leads to novel
and robust algorithms for semi-supervised learning.
Specifically we consider the problem of learning "what
good classifiers are like"
(i.e. predictive structures) from multiple learning
We call this problem structural learning.
In the Bayesian framework, structural learning can be
viewed as hierarchical
modeling, and its simplest form covers the well-known
In this talk, I will give a more flexible and
realistic formulation of the problem
under the standard machine learning framework, in
which rigorous theoretical analysis
can be carried out.
I will then present a specific learning formulation
which can be analyzed in our framework.
The proposed formulation can be solved by an iterative
Finally I will illustrate how to apply this algorithm
to semi-supervised learning.
The idea is to discover the common predictive
structure shared by many automatically generated
auxiliary classification problems using unlabeled
data, and then employ this structure
to improve performance on target problems.
Experiments will be presented to demonstrate the
effectiveness of the proposed method
in the setting of semi-supervised learning.
Joint work with Rie Ando.
| contact email@example.com