Discriminative learning can succeed where generative learning fails
Information Processing Letters, vol. 103(4)
(2007), pp. 131-135
Philip M. Long, Rocco A. Servedio, Hans Ulrich Simon
Generative algorithms for learning classifiers use training data to separately
estimate a probability model for each class. New items are classified by comparing
their probabilities under these models. In contrast, discriminative learning
algorithms try to find classifiers that perform well on all the training data.
We show that there is a learning problem that can be solved by a discriminative
learning algorithm, but not by any generative learning algorithm. This statement
is formalized using a framework inspired by previous work of Goldberg.