In real sequence labeling tasks, statistics of many higher order features are not sufficient due to the training data sparseness, very few of them are useful. We describe Sparse H...
The promise of unsupervised learning methods lies in their potential to use vast amounts of unlabeled data to learn complex, highly nonlinear models with millions of free paramete...
This paper presents a theoretical framework for ranking, and demonstrates how to perform generalization analysis of listwise ranking algorithms using the framework. Many learning-...
Graph based semi-supervised learning (SSL) methods play an increasingly important role in practical machine learning systems. A crucial step in graph based SSL methods is the conv...
We consider the problem of selecting a subset of m most informative features where m is the number of required features. This feature selection problem is essentially a combinator...
Zenglin Xu, Rong Jin, Jieping Ye, Michael R. Lyu, ...