Search research 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

Search research articles

Related Experiment Videos

Multidimensional density shaping by sigmoids.

Z Roth¹, Y Baram

¹Adv. Technol. Center, Qualcomm, Haifa.

IEEE Transactions on Neural Networks

|January 1, 1996

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

Association between sex and the developmental morphokinetics of in vitro derived bovine embryos.

Scientific reports·2025

Same author

Entourage effect for phenolic compounds on production and metabolism of mammary epithelial cells.

Heliyon·2022

Same author

55 Morphokinetics of <i>in vitro</i>-derived bovine embryo is associated with the transcriptomic profile of the developed blastocyst.

Reproduction, fertility, and development·2022

Same author

Carryover effect of atrazine and its metabolite-from treated bovine spermatozoa to the embryo's transcriptome†.

Biology of reproduction·2021

Same author

Pistacia lentiscus extract enhances mammary epithelial cells' productivity by modulating their oxidative status.

Scientific reports·2020

Same author

Mono(2-ethylhexyl) phthalate (MEHP) induces transcriptomic alterations in oocytes and their derived blastocysts.

Toxicology·2019

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 method for estimating probability density functions using feedforward neural networks. This approach enhances classification and real-time prediction accuracy for random variables and sequences.

Area of Science:

Machine Learning
Statistical Modeling
Computational Neuroscience

Background:

Estimating probability density functions (PDFs) is crucial for statistical inference and machine learning.
Traditional methods can be computationally intensive or limited in handling complex data distributions.

Purpose of the Study:

To develop a novel method for PDF estimation using feedforward neural networks.
To apply this method to classification and forecasting problems.

Main Methods:

Maximizing the output entropy of a feedforward network of sigmoidal units with respect to input weights to estimate the PDF.
Utilizing Newton's optimization method for recursive estimation of random variables/sequences.
Implementing constrained connectivity for linear estimation and real-time prediction.

Related Experiment Videos

Employing Gaussian nonlinearity for closed-form solutions and parameter initialization.

Main Results:

The proposed method effectively estimates the probability density function of random vectors.
Classification is achieved by selecting the class with the maximal estimated density.
Newton's method provides a recursive estimator for random variables and sequences.
A constrained connectivity structure enables linear, real-time prediction.
Gaussian nonlinearity simplifies parameter estimation and aids optimization.

Conclusions:

The feedforward network approach offers a robust method for PDF estimation.
The technique is applicable to both classification and forecasting tasks.
Optimized network structures and nonlinearities improve efficiency and performance.