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

Oscillatory networks: pattern recognition without a superposition catastrophe.

Thomas Burwick1

  • 1Institut für Neuroinformatik, Ruhr-Universität Bochum, 44306 Bochum, Germany. thomas.burwick@neuroinformatik.rub.de

Neural Computation
|December 28, 2005
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

Recurrent processing improves occluded object recognition and gives rise to perceptual hysteresis.

Journal of vision·2021
Same author

Few-shot learning in deep networks through global prototyping.

Neural networks : the official journal of the International Neural Network Society·2017
Same author

Selective Interareal Synchronization through Gamma Frequency Differences and Slower-Rhythm Gamma Phase Reset.

Neural computation·2016
Same author

The binding problem.

Wiley interdisciplinary reviews. Cognitive science·2015
Same author

Attentional Bias Through Oscillatory Coherence Between Excitatory Activity and Inhibitory Minima.

Neural computation·2015
Same author

Temporal coding: the relevance of classical activity, its relation to pattern frequency bands, and a remark on recoding of excitatory drive into phase shifts.

Bio Systems·2008
Same journal

A Model-Free Reinforcement Learning Implementation of Decision Making Under Uncertainty by Sequential Sampling.

Neural computation·2026
Same journal

DROP: Distributional and Regular Optimism and Pessimism for Reinforcement Learning.

Neural computation·2026
Same journal

Hierarchical Active Inference Using Successor Representations.

Neural computation·2026
Same journal

W-Kernel and Its Principal Space for Frequentist Evaluation of Bayesian Estimators.

Neural computation·2026
Same journal

A Hidden Markov Model-Inspired Sequence Classification Method for Hyperdimensional Computing.

Neural computation·2026
Same journal

Sparse Graphical Modeling for Electrophysiological Phase-Based Connectivity Using Circular Statistics.

Neural computation·2026
See all related articles

This study presents an oscillatory network model that avoids the superposition catastrophe in pattern recognition. The model uses phase dynamics to allow nonoverlapping patterns to coexist and manage overlapping ones, solving a key challenge in neural networks.

Area of Science:

  • Computational Neuroscience
  • Artificial Intelligence
  • Network Dynamics

Background:

  • The superposition catastrophe poses a significant challenge in pattern recognition, where multiple patterns interfere, leading to degraded retrieval.
  • Classical network models often struggle to simultaneously represent and retrieve overlapping patterns without interference.

Purpose of the Study:

  • To propose and validate an oscillatory network model capable of overcoming the superposition catastrophe.
  • To demonstrate how phase dynamics can be leveraged to achieve robust pattern recognition in complex network architectures.

Main Methods:

  • Developed a novel oscillatory network model integrating classical network principles with phase dynamics.
  • Ensured on/off states correspond to high/low phase velocities for pattern representation.

Related Experiment Videos

  • Employed competition mechanisms to resolve interference for overlapping patterns.

    • Main Results:

    • Successfully demonstrated that nonoverlapping patterns can be simultaneously active with distinct phases.
    • Showcased the model's ability to reduce coherence to a subset of patterns when overlaps occur.
    • Validated the model's capacity to avoid the superposition catastrophe in pattern recognition tasks.

    Conclusions:

    • The proposed oscillatory network model effectively resolves the superposition problem in pattern recognition.
    • Phase dynamics offer a viable mechanism for enhancing the capacity and robustness of neural network models.