Parallel Distributed Processing theory in the age of deep networks

Research output: Contribution to journalArticle (Academic Journal)peer-review

12 Citations (Scopus)
333 Downloads (Pure)

Abstract

Parallel Distributed Processing (PDP) models in psychology are the precursors of deep networks used in computer science. However, only PDP models are associated with two core psychological claims, namely, that all knowledge is coded in a distributed format, and cognition is mediated by non-symbolic computations. These claims have long been debated within cognitive science, and recent work with deep networks speaks to this debate. Specifically, single-unit recordings show that deep networks learn units that respond selectively to meaningful categories, and researchers are finding that deep networks need to be supplemented with symbolic systems in order to perform some tasks. Given the close links between PDP and deep networks, it is surprising that research with deep networks is challenging PDP theory.
Original languageEnglish
Pages (from-to)950-961
Number of pages12
JournalTrends in Cognitive Sciences
Volume21
Issue number12
Early online date31 Oct 2017
DOIs
Publication statusPublished - 1 Dec 2017

Structured keywords

  • Language
  • Cognitive Science

Keywords

  • grandmother cell
  • localist representation
  • distributed representation
  • symbolic representation
  • deep neural network

Fingerprint Dive into the research topics of 'Parallel Distributed Processing theory in the age of deep networks'. Together they form a unique fingerprint.

Cite this