Jove
Visualize
Contact Us
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

Related Experiment Videos

Generalization properties of modular networks: implementing the parity function.

L Franco1, S A Cannas

  • 1Condensed Matter Group, Facultad de Matemática, Astronomía y Física, Universidad Nacional de Córdoba, Córdoba, Argentina. lfranco@sissa.it

IEEE Transactions on Neural Networks
|February 6, 2008
PubMed
Summary
This summary is machine-generated.

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

Evidence for the Collective Nature of Radial Flow in Pb+Pb Collisions with the ATLAS Detector.

Physical review letters·2026
Same author

Evidence for the Dimuon Decay of the Higgs Boson in pp Collisions with the ATLAS Detector.

Physical review letters·2025
Same author

Evidence for Longitudinally Polarized W Bosons in the Electroweak Production of Same-Sign W Boson Pairs in Association with Two Jets in pp Collisions at sqrt[s]=13  TeV with the ATLAS Detector.

Physical review letters·2025
Same author

Observation of tt[over ¯] Production in Pb+Pb Collisions at sqrt[s_{NN}]=5.02  TeV with the ATLAS Detector.

Physical review letters·2025
Same author

Search for Dark Matter Produced in Association with a Dark Higgs Boson in the bb[over ¯] Final State Using pp Collisions at sqrt[s]=13  TeV with the ATLAS Detector.

Physical review letters·2025
Same author

Search for Magnetic Monopole Pair Production in Ultraperipheral Pb+Pb Collisions at sqrt[s_{NN}]=5.36  TeV with the ATLAS Detector at the LHC.

Physical review letters·2025
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

Modular network architectures enhance the generalization ability of Boolean parity functions. Increased modularity improves learning algorithm performance, as shown by analytical and simulation results.

Area of Science:

  • Computational neuroscience
  • Machine learning theory

Background:

  • The parity function is a benchmark for evaluating machine learning algorithms due to its simplicity and complexity.
  • Modular architectures offer a structured approach to implementing complex functions like parity.

Purpose of the Study:

  • To investigate how modular network architectures affect the generalization ability of parity function implementations.
  • To analyze the impact of network modularity and example selection on learning performance.

Main Methods:

  • Construction of a family of modular neural network architectures characterized by maximum fan-in.
  • Analytical computation of the minimum examples required for perfect generalization.
  • Numerical simulations to assess generalization ability and compare learning curves.

Related Experiment Videos

  • Comparison of random example selection versus algorithm-guided selection.

    • Main Results:

    • Generalization ability of modular networks systematically improves with increased modularity.
    • Network modularity is a key factor in achieving better learning performance.
    • Algorithm-guided example selection demonstrates a significant influence on the emergence of generalization ability.

    Conclusions:

    • Modular architectures are effective for implementing the parity function and improving generalization.
    • The degree of modularity directly correlates with enhanced learning performance.
    • Strategic example selection is crucial for optimizing the generalization capabilities of learning algorithms.