Zobrazeno 1 - 10 of 225 pro vyhledávání: '"R. Urbanczik"'
1
Enumerating constrained elementary flux vectors of metabolic networks
Autor: R. Urbanczik
Publikováno v: IET Systems Biology. 1:274-279
The concept of elementary vector is generalised to the case where the steady-state space of the metabolic network is not a flux cone but is a general polyhedron due to further inhomogeneous constraints on the flows through some of the reactions. On o
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2777e1fb6be7183319ab5798f89d972c
https://doi.org/10.1049/iet-syb:20060073
Zobrazit plný text záznamu
2
Training multilayer perceptrons by principal component analysis
Autor: R. Urbanczik, Christoph Bunzmann, Michael Biehl
Publikováno v: Physica A: Statistical Mechanics and its Applications. 302:56-63
We present a training algorithm for multilayer perceptrons which relates to the technique of principal component analysis. The latter is performed with respect to a correlation matrix which is computed from the example inputs and their target outputs
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::c00239532905f0880b1d0b5bff294619
https://doi.org/10.1016/s0378-4371(01)00440-x
Zobrazit plný text záznamu
3
Support vector machines learning noisy polynomial rules
Autor: Manfred Opper, R. Urbanczik
Publikováno v: Physica A: Statistical Mechanics and its Applications. 302:110-118
Using statistical physics, we study support vector machines (SVMs) learning noisy target rules in cases when the optimal predictor is a polynomial of the inputs. If the kernel of the SVM has sufficiently high order or is transcendental, the scale of
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::f27b5be73c7636c48e805f5c4a81bd4e
https://doi.org/10.1016/s0378-4371(01)00446-0
Zobrazit plný text záznamu
4
Statistical mechanics of mutual information maximization
Autor: R. Urbanczik
Publikováno v: Europhysics Letters (EPL). 49:685-691
An unsupervised learning procedure based on maximizing the mutual information between the outputs of two networks receiving different but statistically dependent inputs is analyzed (Becker S. and Hinton G., Nature, 355 (1992) 161). By exploiting a fo
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::82a09b8a76397de3f7585a2831eeda41
https://doi.org/10.1209/epl/i2000-00205-7
Zobrazit plný text záznamu
5
Multilayer perceptrons may learn simple rules quickly
Autor: R. Urbanczik
Publikováno v: Physical Review E. 58:2298-2301
Zero-temperature Gibbs learning is considered for a connected committee machine with $K$ hidden units. For large $K,$ the scale of the learning curve strongly depends on the target rule. When learning a perceptron, the sample size $P$ needed for opti
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::f9240a18a4d6af0b471eabaf8a0de8d3
https://doi.org/10.1103/physreve.58.2298
Zobrazit plný text záznamu
6
Storage capacity of the fully-connected committee machine
Autor: R Urbanczik
Publikováno v: Journal of Physics A: Mathematical and General. 30:L387-L392
The storage capacity, that is the number of patterns which can be stored per weight, is calculated for the fully-connected committee machine with real couplings and K hidden units from the vanishing of the entropy of the internal representations, and
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::e8ffd74219e913452d4c6d3e3f395990
https://doi.org/10.1088/0305-4470/30/11/007
Zobrazit plný text záznamu
8
Learning in a large committee machine: worst case and average case
Autor: R. Urbanczik
Publikováno v: Europhysics Letters (EPL). 35:553-558
Learning of realizable rules is studied for tree committee machines with continuous weights. No nontrivial upper bound exists for the generalization error of consistent students as the number of hidden units K increases. However, numerical considerat
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::29e985fcef2b2ce8ef98810d7327905a
https://doi.org/10.1209/epl/i1996-00149-4
Zobrazit plný text záznamu
9
A Large Committee Machine Learning Noisy Rules
Autor: R. Urbanczik
Publikováno v: Neural Computation. 8:1267-1276
Statistical mechanics is used to study generalization in a tree committee machine with K hidden units and continuous weights trained on examples generated by a teacher of the same structure but corrupted by noise. The corruption is due to additive ga
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::956f87cf196a63cdcb9fbabb87885566
https://doi.org/10.1162/neco.1996.8.6.1267
Zobrazit plný text záznamu
10
A fully connected committee machine learning unrealizable rules
Autor: R Urbanczik
Publikováno v: Journal of Physics A: Mathematical and General. 28:7097-7104
We study generalization in a large fully connected committee machine with continuous weights trained on patterns with outputs generated by a teacher of the same structure but corrupted by noise. The corruption is due to additive Gaussian noise applie
Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::2efa1c4f25fcccf1eca4938fc3d51a48
https://doi.org/10.1088/0305-4470/28/24/010
Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání

Omezení vyhledávání
Plný text Recenzováno Digitální knihovna AV ČR
Zdroje