We investigate the effectiveness of GP-generated intelligent structures in classification tasks. Specifically, we present and use four context-free grammars to describe (1) decisi...
Following Tesauro’s work on TD-Gammon, we used a 4000 parameter feed-forward neural network to develop a competitive backgammon evaluation function. Play proceeds by a roll of t...
This work aims at defining an extension of a competitive method for matching correspondences in stereoscopic image analysis. The method we extended was proposed by Venkatesh, Y.V....
A number of Russian verbs lack 1sg nonpast forms. These paradigmatic gaps are puzzling because they seemingly contradict the highly productive nature of inflectional systems. We m...
Robert Daland, Andrea D. Sims, Janet Pierrehumbert
We propose a probabilistic graphical model to represent weakly annotated images1 . This model is used to classify images and automatically extend existing annotations to new image...