Compositional Connectionism in Cognitive Science
Papers from the 2004 AAAI Fall Symposium
Simon D. Levy and Ross Gayler, Program Cochairs
Technical Report FS-04-03. Published by The AAAI Press, Menlo Park, California
This technical report is also available in book and CD format.
Please Note: Abstracts are linked to individual titles, and will appear in a separate browser window. Full-text versions of the papers are linked to the abstract text. Access to full text may be restricted to AAAI members. PDF file sizes may be large!
Contents
Preface / 97
Simon D. Levy and Ross Gayler
From Wolves Hunting Elk to Rubik’s Cubes: Are the Cortices Composition/Decomposition Engines? / 1
David Arathorn
When Compositionality Fails to Predict Systematicity / 6
Reinhard Blutner, Petra Hendriks, Helen de Hoop, and Oren Schwartz
Context-free versus Context-Dependent Constituency Relations: A False Dichotomy / 12
Francisco Calvo Garzón
Learning Context Sensitive Logical Inference in a Neurobiolobical Simulation / 17
Chris Eliasmith
Scaling Connectionist Compositional Representations / 20
John C. Flackett, John Tait, and Guy Littlefair
Cloning Composition and Logical Inferences in Neural Networks Using Variable-Free Logic / 25
Helmar Gust and Kai-Uwe Kühnberger
A Solution to the Binding Problem for Compositional Connectionism / 31
John E. Hummel, Keith J. Holyoak, Collin Green, Leonidas A. A. Doumas, Derek Devnich, Aniket Kittur, and Donald J. Kalar
Using Simple Recurrent Networks to Learn Fixed-Length Representations of Variable-Length Strings / 35
Christopher T. Kello, Daragh E. Sibley, and Andrew Colombi
Recurrent Representation Reinterpreted / 40
David Landy
Generating Semantic Graphs through Self-Organization / 44
Marshall R. Mayberry III and Matthew W. Crocker
On-Line Learning of Predictive Compositional Hierarchies by Hebbian Chunking / 50
Karl Pfleger
Compositionality in a Knowledge-Based Constructive Learner / 54
François Rivest and Thomas R. Shultz
Where Does Compositionality Come From? / 59
Mark Steedman
On the Relationship between Symbolic and Neural Computation / 63
Whitney Tabor and Dalia Terhesiu
On Early Stages of Learning in Connectionist Models with Feedback Connections / 69
Peter Tino and Barbara Hammer
A Neural Model of Compositional Sentence Structures / 72
Frank van der Velde
Implementing the (De-)Composition of Concepts: Oscillatory Networks, Coherency Chains and Hierarchical Binding / 76
Markus Werning and Alexander Maye
Geometric Ordering of Concepts, Logical Disjunction, and Learning by Induction / 82
Dominic Widdows and Michael Higgins