Maximum Entropy Discrimination

Title: en_US Maximum Entropy Discrimination
Creator: en_US Jaakkola, Tommi; en_US Meila, Marina; en_US Jebara, Tony
Date: 2004-10-20T20:29:28Z
Date Available: 2004-10-20T20:29:28Z
Date Issued: en_US 1999-12-01
Identifier: en_US AITR-1668
uri: http://hdl.handle.net/1721.1/7089
Abstract: en_US We present a general framework for discriminative estimation based on the maximum entropy principle and its extensions. All calculations involve distributions over structures and/or parameters rather than specific settings and reduce to relative entropy projections. This holds even when the data is not separable within the chosen parametric class, in the context of anomaly detection rather than classification, or when the labels in the training set are uncertain or incomplete. Support vector machines are naturally subsumed under this class and we provide several extensions. We are also able to estimate exactly and efficiently discriminative distributions over tree structures of class-conditional models within this framework. Preliminary experimental results are indicative of the potential in these techniques.
Extent: 6420262 bytes; 1702298 bytes
Format: application/postscript; application/pdf
Language: en_US
Relation: en_US AITR-1668