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

Performance enhancement using nonlinear preprocessing.

T Chow1, C T Leung

  • 1Dept. of Electron. Eng., City Polytech. of Hong Kong, Kowloon.

IEEE Transactions on Neural Networks
|January 1, 1996
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

Differentiation of patient-specific void urine-derived human induced pluripotent stem cells to fibroblasts and skeletal muscle myocytes.

Scientific reports·2023
Same author

Poster - Thurs Eve-36: Use of multileaf collimator as a replacement of physical missing tissue compensator.

Medical physics·2017
Same author

Poster - Thurs Eve-34: Extended CT-range in RT planning of pelvic cancer treatment in presence of hip replacements.

Medical physics·2017
Same author

In-vivo imaging of grey and white matter neuroinflammation in Alzheimer's disease: a positron emission tomography study with a novel radioligand, [18F]-FEPPA.

Molecular psychiatry·2015
Same author

A neural-based crowd estimation by hybrid global learning algorithm.

IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2008
Same author

Erythropoietin receptor expression in biopsy specimens from patients with uterine cervix squamous cell carcinoma.

International journal of gynecological cancer : official journal of the International Gynecological Cancer Society·2006
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 novel nonlinear preprocessing method enhances network performance by improving input-output vector distribution, leading to easier optimization. This technique significantly reduces errors, achieving up to a 98% decrease in test set error.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Data Science

Background:

  • Network performance is often limited by input-output vector distributions.
  • Optimization of complex networks can be challenging due to data characteristics.

Purpose of the Study:

  • To introduce and evaluate a nonlinear preprocessing method for enhancing network output performance.
  • To demonstrate the method's ability to improve network optimization and reduce errors.

Main Methods:

  • A nonlinear preprocessing technique was developed to modify input and output vector distributions.
  • The method aims to increase the orthogonality between input and output variables.
  • Performance was evaluated using time-series prediction applications.

Related Experiment Videos

Main Results:

  • The nonlinear preprocessing method effectively redistributes data distributions.
  • Input and output variables become more orthogonal, facilitating network optimization.
  • Significant reductions in test set error, up to 98%, were observed in evaluated examples.

Conclusions:

  • Nonlinear preprocessing is a viable strategy to enhance network performance.
  • The proposed method offers a significant improvement in predictive accuracy.
  • The technique shows promise for various time-series prediction tasks.