Sciweavers

986 search results - page 47 / 198
» Zero-data Learning of New Tasks
Sort
View
NPL
2006
90views more  NPL 2006»
14 years 9 months ago
Hierarchical Incremental Class Learning with Reduced Pattern Training
Hierarchical Incremental Class Learning (HICL) is a new task decomposition method that addresses the pattern classification problem. HICL is proven to be a good classifier but clos...
Sheng Uei Guan, Chunyu Bao, Ru-Tian Sun
NAACL
2010
14 years 7 months ago
Painless Unsupervised Learning with Features
We show how features can easily be added to standard generative models for unsupervised learning, without requiring complex new training methods. In particular, each component mul...
Taylor Berg-Kirkpatrick, Alexandre Bouchard-C&ocir...
JMLR
2012
13 years 8 days ago
Bayesian Comparison of Machine Learning Algorithms on Single and Multiple Datasets
We propose a new method for comparing learning algorithms on multiple tasks which is based on a novel non-parametric test that we call the Poisson binomial test. The key aspect of...
Alexandre Lacoste, François Laviolette, Mar...
IJCNN
2000
IEEE
15 years 2 months ago
Bias Learning, Knowledge Sharing
—Biasing properly the hypothesis space of a learner has been shown to improve generalization performance. Methods for achieving this goal have been proposed, that range from desi...
Joumana Ghosn, Yoshua Bengio
EMNLP
2010
14 years 7 months ago
Efficient Graph-Based Semi-Supervised Learning of Structured Tagging Models
We describe a new scalable algorithm for semi-supervised training of conditional random fields (CRF) and its application to partof-speech (POS) tagging. The algorithm uses a simil...
Amarnag Subramanya, Slav Petrov, Fernando Pereira