Jove
Visualize
Contact Us

Related Experiment Videos

A connectionist computational model for epistemic and temporal reasoning.

Artur S d'Avila Garcez1, Luís C Lamb

  • 1Department of Computing, City University, London EC1V 0HB, UK. aag@soi.city.ac.uk

Neural Computation
|June 13, 2006
PubMed
Summary

Neural-symbolic AI systems can now learn and reason with complex temporal and modal logic. This advance enables artificial intelligence to better handle knowledge representation and learning over time.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A Bayesian predictive analytics model for improving long range epidemic forecasting during an infection wave.

Healthcare analytics (New York, N.Y.)·2023
Same author

Deep Logic Networks: Inserting and Extracting Knowledge From Deep Belief Networks.

IEEE transactions on neural networks and learning systems·2016
Same author

Three-dimensional protein structure prediction: Methods and computational strategies.

Computational biology and chemistry·2014
Same author

Learning and representing temporal knowledge in recurrent networks.

IEEE transactions on neural networks·2011
Same author

Collective Poisson process with periodic rates: applications in physics from micro-to nanodevices.

Philosophical transactions. Series A, Mathematical, physical, and engineering sciences·2010
Same author

Statistical fluctuations in population bargaining in the ultimatum game: static and evolutionary aspects.

Journal of theoretical biology·2009
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Area of Science:

  • Artificial Intelligence
  • Computational Logic
  • Machine Learning

Background:

  • Bridging connectionist and symbolic AI paradigms is crucial for advanced AI systems.
  • Neural-symbolic systems integrate background knowledge with data learning for enhanced effectiveness.
  • Previous neural-symbolic systems had limitations in representing and reasoning with expressive nonclassical logics.

Purpose of the Study:

  • To demonstrate the effective computation of nonclassical logics, specifically temporal and epistemic reasoning, by artificial neural networks.
  • To introduce a novel language, connectionist temporal logic of knowledge (CTLK), for neural-symbolic systems.
  • To apply CTLK to complex reasoning problems like the muddy children puzzle.

Main Methods:

  • Development of the connectionist temporal logic of knowledge (CTLK) language.

Related Experiment Videos

  • Design of a temporal algorithm to translate CTLK theories into ensembles of neural networks.
  • Verification of the translation algorithm's correctness.
  • Application of CTLK and neural networks to solve the muddy children puzzle.

    • Main Results:

    • Nonclassical logics, including temporal and epistemic reasoning, are effectively computed by artificial neural networks.
    • The CTLK language and translation algorithm provide a correct method for integrating these logics into neural networks.
    • A complete solution to the muddy children puzzle was achieved using simple neural networks.

    Conclusions:

    • Neural-symbolic systems can effectively represent, reason, and learn expressive nonclassical logics.
    • CTLK offers a powerful framework for AI systems to reason about knowledge evolution and acquisition over time.
    • This research advances AI capabilities in distributed knowledge representation and complex reasoning tasks.