Manual generation of training examples for supervised learning is an expensive process. One way to reduce this cost is to produce training instances that are highly informative. T...
Justus H. Piater, Edward M. Riseman, Paul E. Utgof...
A fundamental assumption for any machine learning task is to have training and test data instances drawn from the same distribution while having a sufficiently large number of tra...
We discuss a simple sparse linear problem that is hard to learn with any algorithm that uses a linear combination of the training instances as its weight vector. The hardness holds...
A major difficulty of supervised approaches for text classification is that they require a great number of training instances in order to construct an accurate classifier. This pap...
In this paper, we propose the "Democratic Classifier", a simple, democracy-inspired patternbased classification algorithm that uses very short patterns for classificatio...
Support Vector Machines (SVMs) suffer from an O(n2 ) training cost, where n denotes the number of training instances. In this paper, we propose an algorithm to select boundary ins...