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 recurrent log-linearized Gaussian mixture network.

T Tsuji1, Nan Bu, O Fukuda

  • 1Dept. of Artificial Complex Syst. Eng., Hiroshima Univ., Higashi, Japan.

IEEE Transactions on Neural Networks
|February 2, 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

Reversible interactions between plasminogen activators and plasminogen activator inhibitor-1.

Biochimica et biophysica acta·1992
Same author

[Transjugular intrahepatic portosystemic stent shunt (TIPSS) with use of Rösch-Uchida transjugular liver access set--evaluation by CT and its clinical application].

Nihon Igaku Hoshasen Gakkai zasshi. Nippon acta radiologica·1992
Same author

[Relationship between health services for maternal and child health and working hours of public health nurses].

[Nihon koshu eisei zasshi] Japanese journal of public health·1992
Same author

[Coronet status of digital radiography in chest medicine].

Nihon Kyobu Shikkan Gakkai zasshi·1992
Same author

[Gianturco expandable metallic stents in the treatment of tumor thrombus in portal vein--preliminary clinical experience].

Nihon Igaku Hoshasen Gakkai zasshi. Nippon acta radiologica·1992
Same author

Decrease in heart mitochondrial creatine kinase activity due to oxygen free radicals.

Biochimica et biophysica acta·1992
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

A new recurrent log-linearized Gaussian mixture network (R-LLGMN) effectively classifies time series data. This novel neural network, inspired by hidden Markov models, shows strong performance on both artificial and real-world biological signals like EEG.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Signal Processing

Background:

  • Time series context is crucial for machine learning classification.
  • Dynamic characteristics are often the sole basis for classification.
  • Hidden Markov Models (HMMs) are established in speech recognition.

Purpose of the Study:

  • Introduce a novel neural network for time series classification.
  • Leverage temporal information using recurrent connections.
  • Compare the proposed network against traditional HMM classifiers.

Main Methods:

  • Developed a recurrent log-linearized Gaussian mixture network (R-LLGMN).
  • Structured the network based on Hidden Markov Model (HMM) principles.
  • Incorporated recurrent connections to utilize temporal dynamics.

Related Experiment Videos

Main Results:

  • R-LLGMN demonstrated successful classification of artificial time series data.
  • The network effectively classified real biological data, specifically EEG signals.
  • Performance was comparable to traditional HMM estimators in simulation experiments.

Conclusions:

  • R-LLGMN is a viable and effective model for time series classification.
  • The network successfully integrates temporal context for improved classification.
  • Applicable to both synthetic and complex biological signal analysis.