Learning of Compositional Hierarchies for the modeling of context effects

From Tetherless World Wiki

Jump to: navigation, search

Citation: Karl Pfleger and Barbara Hayes-Roth. (1998) Learning of Compositional Hierarchies for the modeling of context effects. In KSL-98-04, January,1998.

Publication techreport ( Edit )
type Technical Report
bibtype techreport
Bibtex basics
author Karl Pfleger and Barbara Hayes-Roth
title Learning of Compositional Hierarchies for the modeling of context effects
number KSL-98-04
institution Knowledge Systems, AI Laboratory
year 1998
month January
Bibtex more
Access Paper
abstract Compositional, or part-whole, hierarchies underlie many forms of data, and representations involving these structures lie at the heart of much of the work in Artificial Intelligence and Cognitive Science. However, despite their prevalence, general methods for learning such structures from data are scarce. This paper presents a learning and prediction system that learns compositional hierarchies and uses them to mediate context effects in making predictions. The model is a hybrid system based on an early psychological neural network system, the Interactive Activation model of context effects in letter perception, and an elegant new symbolic hierarchy-generation algorithm called Sequitur. The composite system overcomes an important limitation in each of its parents.

KSL Technical Report ID: KSL-98-04
Facts about Learning of Compositional Hierarchies for the modeling of context effectsRDF feed
Abstract Compositional, or part-whole, hierarchies … Compositional, or part-whole, hierarchies underlie many forms of data, and representations involving these structures lie at the heart of much of the work in Artificial Intelligence and Cognitive Science. However, despite their prevalence, general methods for learning such structures from data are scarce. This paper presents a learning and prediction system that learns compositional hierarchies and uses them to mediate context effects in making predictions. The model is a hybrid system based on an early psychological neural network system, the Interactive Activation model of context effects in letter perception, and an elegant new symbolic hierarchy-generation algorithm called Sequitur. The composite system overcomes an important limitation in each of its parents. portant limitation in each of its parents.
Author Karl Pfleger and Barbara Hayes-Roth  +
Bibtype techreport  +
Has author Karl Pfleger and Barbara Hayes-Roth  +
Has identifier KSL-98-04  +
Has publishing details January,1998  +
Has title Learning of Compositional Hierarchies for the modeling of context effects  +
Has where published KSL-98-04  +
Has year 1998  +
Institution Knowledge Systems, AI Laboratory  +
Ksl tr id KSL-98-04  +
Month January  +
Number KSL-98-04  +
Process note NO  +
Title Learning of Compositional Hierarchies for the modeling of context effects  +
Year 1998  +
Retrieved from "http://tw.rpi.edu/wiki/Learning_of_Compositional_Hierarchies_for_the_modeling_of_context_effects"
Personal tools