One or more hypotheses may be induced from any set of exemplars and non exemplars. Incremental Modification of Hypothesis Fragments (IMHF) is a new algorithm for processing instances of concepts incrementally, discovering a consistent hypothesis after presentation of each example. / A modular connectionist approach using the back-propagation learning algorithm was taken to implement the IMHF algorithm. A shell called Parallel Unconnected Neural Networks (PUNN) was developed to give back-propagation the additional power of modularity and provided for the needed complexity to model IMHF. The PUNN implementation of the IMHF algorithm yielded a model of human induction of hypotheses from examples.
Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.60100 |
Date | January 1990 |
Creators | Strigler, David |
Publisher | McGill University |
Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
Language | English |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Format | application/pdf |
Coverage | Master of Science (School of Computer Science.) |
Rights | All items in eScholarship@McGill are protected by copyright with all rights reserved unless otherwise indicated. |
Relation | alephsysno: 001238689, proquestno: AAIMM67846, Theses scanned by UMI/ProQuest. |
Page generated in 0.0015 seconds