We propose a new family of latent variable models called max-margin min-entropy (m3e) models, which define a distribution over the output and the hidden variables conditioned on ...
Kevin Miller, M. Pawan Kumar, Benjamin Packer, Dan...
Traditional machine learning makes a basic assumption: the training and test data should be under the same distribution. However, in many cases, this identicaldistribution assumpt...
State machines are considered a very general means of expressing computations in an implementation-independent way. There are also ways to extend the general state machine framewor...
With a few exceptions, discriminative training in statistical machine translation (SMT) has been content with tuning weights for large feature sets on small development data. Evid...
In this paper an instructional framework is proposed for supporting personalised learning in the context of webbased adaptive educational hypermedia systems. A learning-focused ap...