A computational holographic model of memory for abstract associations

Matthew A. Kelly, Robert L. West

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

How do humans learn the syntax and semantics of words from language experience? How does the mind discover abstract relationships between concepts? Computational models of distributional semantics can analyze a corpus to derive representations of word meanings in terms of each word’s relationship to all other words in the corpus. While these models are sensitive to topic (e.g., tiger and stripes) and synonymy (e.g., soar and fly), the models have limited sensitivity to part of speech (e.g., book and shirt are both nouns). By augmenting a holographic model of semantic memory with additional layers of representations, we demonstrate that sensitivity to syntax relies on exploiting higher-order associations between words. Our hierarchical holographic memory model bridges the gap between models of distributional semantics and unsupervised part-of-speech induction algorithms, providing evidence that semantics and syntax exist on a continuum and emerge from a unitary cognitive system.

Original languageEnglish (US)
Title of host publicationProceedings of ICCM 2016 - 14th International Conference on Cognitive Modeling
EditorsDavid Reitter, Frank E. Ritter
PublisherThe Pennsylvania State University
Pages279-281
Number of pages3
ISBN (Electronic)9780998508207
StatePublished - 2016
Event14th International Conference on Cognitive Modeling, ICCM 2016 - University Park, United States
Duration: Aug 3 2016Aug 6 2016

Publication series

NameProceedings of ICCM 2016 - 14th International Conference on Cognitive Modeling

Conference

Conference14th International Conference on Cognitive Modeling, ICCM 2016
CountryUnited States
CityUniversity Park
Period8/3/168/6/16

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Modeling and Simulation

Fingerprint Dive into the research topics of 'A computational holographic model of memory for abstract associations'. Together they form a unique fingerprint.

Cite this