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

Processing directed acyclic graphs with recursive neural networks.

M Bianchini1, M Gori, F Scarselli

  • 1Department of Ingegneria dell'Informazione, Università di Siena, Siena 53100, Italy. monica@dii.unisi.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

Primary visual cortex BOLD responses to relative localization of sounds at 7T.

iScience·2026
Same author

New insights into aging in LiNiO<sub>2</sub> cathodes from high resolution paramagnetic NMR spectroscopy.

Chemical communications (Cambridge, England)·2024
Same author

Gender oncology: recommendations and consensus of the Italian Association of Medical Oncology (AIOM).

ESMO open·2024
Same author

Clinical variables associated with late-onset thrombotic and cardiovascular events, after SARS-CoV-2 infection, in a cohort of patients from the first epidemic wave: an 18-month analysis on the "Surviving-COVID" cohort from Bergamo, Italy.

Frontiers in cardiovascular medicine·2023
Same author

Dr-MUSIC: An Effective Device for Investigating Multisensory Mechanisms during Development with EEG recordings<sup></sup>.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2023
Same author

MultiTab: A Novel Portable Device to Evaluate Multisensory Skills<sup></sup>.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2023
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

Recursive neural networks (RNNs) can now process unordered graphs by relaxing ordering constraints. This novel weight sharing method maintains approximation capabilities for graph processing applications.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Recursive neural networks (RNNs) are designed for graph processing, extending recurrent models for sequences.
  • Existing RNNs are limited to directed acyclic graphs (DAGs) with ordered children, imposing constraints on graph structures.

Purpose of the Study:

  • To relax the ordering constraint in recursive neural networks for broader graph processing applicability.
  • To develop a method that handles graphs with unordered children and low connectivity.

Main Methods:

  • Introduced a novel weight sharing technique to ensure network output independence from arc permutations.
  • Developed an architecture capable of processing directed acyclic graphs (DAGs) with unordered children.

Related Experiment Videos

Main Results:

  • The proposed weight sharing method successfully relaxes the ordering constraint on nodes.
  • The new architecture maintains the approximation capabilities of recursive neural networks.
  • The method is effective for graphs with low connectivity and few outgoing arcs.

Conclusions:

  • The developed recursive neural network architecture offers greater flexibility in graph processing.
  • This advancement enables the application of RNNs to a wider range of graph-structured data without ordering limitations.