Limits...
Asymmetric variate generation via a parameterless dual neural learning algorithm.

Fiori S - Comput Intell Neurosci (2008)

Bottom Line: In a previous work (S.Fiori, 2006), we proposed a random number generator based on a tunable non-linear neural system, whose learning rule is designed on the basis of a cardinal equation from statistics and whose implementation is based on look-up tables (LUTs).The new method proposed here proves easier to implement and relaxes some previous limitations.

View Article: PubMed Central - PubMed

Affiliation: Dipartimento di Elettronica, Intelligenza Artificiale e Telecomunicazioni (DEIT), Università Politecnica delle Marche Via Brecce Bianche, Ancona I-60131, Italy. fiori@deit.univpm.it

ABSTRACT
In a previous work (S. Fiori, 2006), we proposed a random number generator based on a tunable non-linear neural system, whose learning rule is designed on the basis of a cardinal equation from statistics and whose implementation is based on look-up tables (LUTs). The aim of the present manuscript is to improve the above-mentioned random number generation method by changing the learning principle, while retaining the efficient LUT-based implementation. The new method proposed here proves easier to implement and relaxes some previous limitations.

No MeSH data available.


Related in: MedlinePlus

Result of dual neuralsystem adaptation with Gaussian input andgeneralized Gaussian output.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2376097&req=5

fig4: Result of dual neuralsystem adaptation with Gaussian input andgeneralized Gaussian output.

Mentions: The numerical results presented below pertain to values σ = 1, μ = 0.8, and λ = 0.5. The interval of interest for the output variable is set to 𝓎 = [−3 4]. The total number of generated output samples amounts to Q = 68335. The number of points in which the function g(⋅) is computed is N = 1000. The results obtained by running the learning algorithm (7) are shown in Figure 4. The values of the index Δgn show that the fixed-point algorithm may be safely stopped after 5 iterations again. In Figure 4, the histogram estimates (with 50 bins) of the generated Gaussian data and of the generalized Gaussian output may be observed as well.


Asymmetric variate generation via a parameterless dual neural learning algorithm.

Fiori S - Comput Intell Neurosci (2008)

Result of dual neuralsystem adaptation with Gaussian input andgeneralized Gaussian output.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2376097&req=5

fig4: Result of dual neuralsystem adaptation with Gaussian input andgeneralized Gaussian output.
Mentions: The numerical results presented below pertain to values σ = 1, μ = 0.8, and λ = 0.5. The interval of interest for the output variable is set to 𝓎 = [−3 4]. The total number of generated output samples amounts to Q = 68335. The number of points in which the function g(⋅) is computed is N = 1000. The results obtained by running the learning algorithm (7) are shown in Figure 4. The values of the index Δgn show that the fixed-point algorithm may be safely stopped after 5 iterations again. In Figure 4, the histogram estimates (with 50 bins) of the generated Gaussian data and of the generalized Gaussian output may be observed as well.

Bottom Line: In a previous work (S.Fiori, 2006), we proposed a random number generator based on a tunable non-linear neural system, whose learning rule is designed on the basis of a cardinal equation from statistics and whose implementation is based on look-up tables (LUTs).The new method proposed here proves easier to implement and relaxes some previous limitations.

View Article: PubMed Central - PubMed

Affiliation: Dipartimento di Elettronica, Intelligenza Artificiale e Telecomunicazioni (DEIT), Università Politecnica delle Marche Via Brecce Bianche, Ancona I-60131, Italy. fiori@deit.univpm.it

ABSTRACT
In a previous work (S. Fiori, 2006), we proposed a random number generator based on a tunable non-linear neural system, whose learning rule is designed on the basis of a cardinal equation from statistics and whose implementation is based on look-up tables (LUTs). The aim of the present manuscript is to improve the above-mentioned random number generation method by changing the learning principle, while retaining the efficient LUT-based implementation. The new method proposed here proves easier to implement and relaxes some previous limitations.

No MeSH data available.


Related in: MedlinePlus