Stable Mixing of Complete and Incomplete Information
Author(s)
Corduneanu, Adrian; Jaakkola, Tommi
DownloadAIM-2001-030.ps (1.151Mb)
Additional downloads
Metadata
Show full item recordAbstract
An increasing number of parameter estimation tasks involve the use of at least two information sources, one complete but limited, the other abundant but incomplete. Standard algorithms such as EM (or em) used in this context are unfortunately not stable in the sense that they can lead to a dramatic loss of accuracy with the inclusion of incomplete observations. We provide a more controlled solution to this problem through differential equations that govern the evolution of locally optimal solutions (fixed points) as a function of the source weighting. This approach permits us to explicitly identify any critical (bifurcation) points leading to choices unsupported by the available complete data. The approach readily applies to any graphical model in O(n^3) time where n is the number of parameters. We use the naive Bayes model to illustrate these ideas and demonstrate the effectiveness of our approach in the context of text classification problems.
Date issued
2001-11-08Other identifiers
AIM-2001-030
Series/Report no.
AIM-2001-030
Keywords
AI, semi-supervised learning, incomplete data, EM, stable estimation