Abstract
Is it possible to combine the strongly complementary properties of discriminative estimation with generative modeling? Can, for instance, support vector machines and the performance gains they provide be combined elegantly with flexible Bayesian statistics and graphical models? This chapter introduces a novel technique called maximum entropy discrimination (MED), which provides a general formalism for marrying both methods [80].
It is futile to do with more what can be done with fewer 1. William of Ockham , 1280-1349
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer Science+Business Media New York
About this chapter
Cite this chapter
Jebara, T. (2004). Maximum Entropy Discrimination. In: Machine Learning. The International Series in Engineering and Computer Science, vol 755. Springer, Boston, MA. https://doi.org/10.1007/978-1-4419-9011-2_3
Download citation
DOI: https://doi.org/10.1007/978-1-4419-9011-2_3
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-4756-9
Online ISBN: 978-1-4419-9011-2
eBook Packages: Springer Book Archive