More data means less inference: A pseudo-max approach to structured learning

More data means less inference: A pseudo-max approach to structured learning

The problem of learning to predict structured labels is of key importance in many applications. However, for general graph structure both learning and inference in this setting are intractable. Here we show that it is possible to circumvent this difficulty when the input distribution is rich enough...

Full description

Bibliographic Details
Main Authors: Sontag, David (Author), Meshi, Ofer (Author), Jaakkola, Tommi S. (Contributor), Globerson, Amir (Author)
Other Authors: Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory (Contributor), Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science (Contributor)
Format: Article
Language:English
Published: Neural Information Processing Systems Foundation, 2011-07-06T14:56:22Z.
Subjects:
Article
Online Access:Get fulltext
Description
Summary:The problem of learning to predict structured labels is of key importance in many applications. However, for general graph structure both learning and inference in this setting are intractable. Here we show that it is possible to circumvent this difficulty when the input distribution is rich enough via a method similar in spirit to pseudo-likelihood. We show how our new method achieves consistency, and illustrate empirically that it indeed performs as well as exact methods when sufficiently large training sets are used.
United States-Israel Binational Science Foundation (Grant 2008303)
Google (Firm) (Research Grant)
Google (Firm) (PhD Fellowship)

Similar Items