Search research articles

Search research articles

Related Experiment Videos

Object selection based on oscillatory correlation.

¹Department of Computer and Information Science and Center for Cognitive Science, The Ohio State University, 2015 Neil Avenue, Columbus, OH, USA

Neural Networks : the Official Journal of the International Neural Network Society

|March 29, 2003

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 journal

TraNce: Type-aware hypergraph neural network with biological mediators for drug repositioning.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

Decentralized ADMM for factorization-based Low-rank matrix estimation.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

Memristive neuromorphic circuit design inspired by the neural mechanisms of conditioned fear.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

Q-learning based asynchronous Boolean control networks stabilization with data loss.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

New results on prescribed-time synchronization of complex networks via intermittent control.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

Variance-constrained multi-view ensemble broad network for imbalanced data.

Neural networks : the official journal of the International Neural Network Society·2026

See all related articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

This study introduces a novel neural network architecture that preserves spatial relationships for object selection. The new model enhances winner-take-all (WTA) networks for improved sensory and perceptual processing.

Area of Science:

Computational neuroscience
Artificial intelligence
Neural network architectures

Background:

Winner-take-all (WTA) networks are fundamental in unsupervised learning and attentional control.
Traditional WTA networks lack spatial encoding, limiting their use in tasks requiring spatial relation processing.
Existing WTA models struggle with sensory and perceptual data where spatial relationships are crucial.

Purpose of the Study:

To propose a novel neural network architecture that preserves spatial relationships in input features.
To extend winner-take-all (WTA) network capabilities for enhanced sensory and perceptual processing.
To develop a selection network that can identify and isolate salient objects within complex scenes.

Main Methods:

The proposed architecture builds upon Locally Excitatory Globally Inhibitory Oscillator Networks (LEGION) dynamics.

Related Experiment Videos

Incorporation of slow inhibition mechanisms to manage network activity and selection.

Development of a two-stage selection process for efficiency and noise removal.

Application of the network to select the most salient object in gray-level images.

Main Results:

The network successfully selects the largest object in scenes with multiple objects.
The system can be configured to select multiple objects, allowing for temporal alternation.
Analysis demonstrates the speed of object selection.
A two-stage variant shows increased efficiency by integrating selection with parallel noise removal.
A special case without local excitation results in a novel oscillatory WTA.

Conclusions:

The novel architecture effectively maintains spatial relations, overcoming limitations of traditional WTA networks.
The proposed selection network offers enhanced capabilities for salient object identification in images.
The findings contribute to the development of more sophisticated neural networks for complex perceptual tasks.
The research presents a new form of oscillatory WTA through a modified network configuration.