Approximating Bayesian inference with a sparse distributed memory system

Joshua T. Abbott, Jessica B. Hamrick, Thomas L. Griffiths

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

10 Scopus citations

Abstract

Probabilistic models of cognition have enjoyed recent success in explaining how people make inductive inferences. Yet, the difficult computations over structured representations that are often required by these models seem incompatible with the continuous and distributed nature of human minds. To reconcile this issue, and to understand the implications of constraints on probabilistic models, we take the approach of formalizing the mechanisms by which cognitive and neural processes could approximate Bayesian inference. Specifically, we show that an associative memory system using sparse, distributed representations can be reinterpreted as an importance sampler, a Monte Carlo method of approximating Bayesian inference. This capacity is illustrated through two case studies: a simple letter reconstruction task, and the classic problem of property induction. Broadly, our work demonstrates that probabilistic models can be implemented in a practical, distributed manner, and helps bridge the gap between algorithmic- and computational-level models of cognition.

Original languageEnglish (US)
Title of host publicationCooperative Minds
Subtitle of host publicationSocial Interaction and Group Dynamics - Proceedings of the 35th Annual Meeting of the Cognitive Science Society, CogSci 2013
EditorsMarkus Knauff, Natalie Sebanz, Michael Pauen, Ipke Wachsmuth
PublisherThe Cognitive Science Society
Pages1686-1691
Number of pages6
ISBN (Electronic)9780976831891
StatePublished - 2013
Externally publishedYes
Event35th Annual Meeting of the Cognitive Science Society - Cooperative Minds: Social Interaction and Group Dynamics, CogSci 2013 - Berlin, Germany
Duration: Jul 31 2013Aug 3 2013

Publication series

NameCooperative Minds: Social Interaction and Group Dynamics - Proceedings of the 35th Annual Meeting of the Cognitive Science Society, CogSci 2013

Conference

Conference35th Annual Meeting of the Cognitive Science Society - Cooperative Minds: Social Interaction and Group Dynamics, CogSci 2013
Country/TerritoryGermany
CityBerlin
Period7/31/138/3/13

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Science Applications
  • Human-Computer Interaction
  • Cognitive Neuroscience

Keywords

  • associative memory models
  • Bayesian inference
  • importance sampling
  • rational process models
  • sparse distributed memory

Fingerprint

Dive into the research topics of 'Approximating Bayesian inference with a sparse distributed memory system'. Together they form a unique fingerprint.

Cite this