This paper discusses an alternative approach to parameter optimization of well-known prototype-based learning algorithms (minimizing an objective function via gradient search). The proposed approach considers a stochastic optimization called the cross entropy method (CE method). The CE method is use
β¦ LIBER β¦
Application of the cross-entropy method to clustering and vector quantization
β Scribed by Dirk P. Kroese; Reuven Y. Rubinstein; Thomas Taimre
- Publisher
- Springer US
- Year
- 2006
- Tongue
- English
- Weight
- 541 KB
- Volume
- 37
- Category
- Article
- ISSN
- 0925-5001
No coin nor oath required. For personal study only.
π SIMILAR VOLUMES
Application of the cross entropy method
β
Abderrahmane Boubezoul; SΓ©bastien Paris; Mustapha Ouladsine
π
Article
π
2008
π
Elsevier Science
π
English
β 389 KB
The Generalized Cross Entropy Method, wi
β
Zdravko I. Botev; Dirk P. Kroese
π
Article
π
2009
π
Springer US
π
English
β 778 KB
Application of the entropy generation mi
β
Mehdi Baneshi; Khosrow Jafarpur; Mojtaba Mahzoon
π
Article
π
2009
π
Springer
π
English
β 451 KB
Quantization of Coulomb wave vectors and
β
I.-J. Kang; R.L. Kerch
π
Article
π
1970
π
Elsevier Science
π
English
β 135 KB
Application of the Cross-Entropy Method
β
G. Alon; D. P. Kroese; T. Raviv; R. Y. Rubinstein
π
Article
π
2005
π
Springer US
π
English
β 217 KB
Application of the star-product method t
β
Paul Molin
π
Article
π
1992
π
Springer
π
English
β 379 KB
We define a .-product on ~]~3 \_\_ {0} and solve the polarization equation f . C = 0 where C is the Casimir of the coadjoint representation of SO(3). We compute the action of SO(3) on the space of solutions. We then examine the case of non-zero eigenvalues of C, in order to find finite-dimensional r