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

A new recurrent neural-network architecture for visual pattern recognition.

S W Lee1, H H Song

  • 1Dept. of Comput. Sci. and Eng., Korea Univ., Seoul.

IEEE Transactions on Neural Networks
|January 1, 1997
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

[Effects of adipose stem cell-derived exosomes on rat tendon healing and its impact on the periphery neuropeptides expression].

Zhonghua yi xue za zhi·2025
Same author

[Value of contrast - enhanced ultrasonography and acoustic radiation force impulse elastography in identification of boundary range and viability of hepatic alveolar echinococcosis].

Zhongguo xue xi chong bing fang zhi za zhi = Chinese journal of schistosomiasis control·2020
Same author

[Factors associated with smoking cessation attempts in male current smokers in rural area of Shandong province].

Zhonghua liu xing bing xue za zhi = Zhonghua liuxingbingxue zazhi·2016
Same author

A unique surface-initiated property of nanoparticles and application for the synthesis of hybrid organic-inorganic nanoparticles.

Chemical communications (Cambridge, England)·2014
Same author

Comparison of RECIST 1.0 and RECIST 1.1 on computed tomography in patients with metastatic colorectal cancer.

Oncology·2014
Same author

Effects of in ovo injection of carbohydrate solution on small intestine development in domestic pigeons (Columba livia).

Journal of animal science·2013
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

This study introduces a novel recurrent neural network architecture that enhances visual pattern recognition. The new model demonstrates improved discrimination and generalization capabilities compared to existing methods.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Neural Networks

Background:

  • Recurrent neural networks (RNNs) are crucial for sequential data processing.
  • Existing architectures like Jordan's and Elman's RNNs have limitations in discrimination and generalization.
  • Feedforward neural networks lack the temporal processing capabilities of RNNs.

Purpose of the Study:

  • To propose a novel recurrent neural network architecture with enhanced connectivity.
  • To improve the discrimination and generalization powers of neural networks for pattern recognition.
  • To analyze the convergence properties and performance of the proposed architecture.

Main Methods:

  • Developed a new RNN architecture with self-connected output units and full connectivity to other units and hidden layers.

Related Experiment Videos

  • Proved the convergence properties of the associated learning algorithm.
  • Conducted recognition experiments using the Concordia University handwritten numeric database.

    • Main Results:

    • The proposed recurrent neural network architecture demonstrated superior performance in visual pattern recognition.
    • Experimental results confirmed significant improvements in discrimination and generalization powers.
    • The model effectively recognized handwritten numeric characters from an unconstrained database.

    Conclusions:

    • The novel recurrent neural network architecture offers enhanced capabilities for visual pattern recognition tasks.
    • The proposed model surpasses existing RNNs in terms of discrimination and generalization.
    • This architecture holds promise for advancing machine learning applications in pattern recognition.