Zobrazeno 1 - 10
of 19
pro vyhledávání: '"Katsuki Katayama"'
Autor:
Katsuki Katayama
Publikováno v:
Physica A: Statistical Mechanics and its Applications. 361:543-568
We make an estimation of a maximizer of posterior marginals (MPM) as for a multi-valued information symbol by a code-division multiple access (CDMA) demodulator using a multi-valued spreading code sequence within a framework of a Bayesian inference.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ea409845da3214e80d005d5d559fa92e
https://doi.org/10.1016/j.physa.2005.05.041
https://doi.org/10.1016/j.physa.2005.05.041
Autor:
Katsuki Katayama
Publikováno v:
Interdisciplinary Information Sciences. 12:19-31
We investigate a restoration process for a gray-scale pattern given by a snapshot of the Q-states ferromagnetic Husimi–Temperly model within a framework of a Bayesian inference. By using a generating function of path-integral representation, we der
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::0c840631c1d52b3abdd5e27a8ef3f732
https://doi.org/10.4036/iis.2006.19
https://doi.org/10.4036/iis.2006.19
Autor:
Tsuyoshi Horiguchi, Katsuki Katayama
Publikováno v:
Progress of Theoretical Physics Supplement. 157:266-269
Storage capacity as for retrieval of sequences of binary patterns is investigated for a fully connected neural network with Q (> 2)-states. By using a generating-function method of path-integral representation, we find that the network with Q-states
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::33316aa322d01413e1be5ad407f910f8
https://doi.org/10.1143/ptps.157.266
https://doi.org/10.1143/ptps.157.266
Publikováno v:
Biological Cybernetics. 91:315-325
We propose a mathematical model of selective visual attention using a two-layered neural network with neurons described by the Hodgkin---Huxley equation in order to investigate part of the assumption proposed by Desimone and Duncan. The neural networ
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c18dbec7e57a66c7af0ac093b417cd65
https://doi.org/10.1007/s00422-004-0504-4
https://doi.org/10.1007/s00422-004-0504-4
Publikováno v:
Physica A: Statistical Mechanics and its Applications. 322:531-545
We propose a model for a system with middle temporal neurons and medial superior temporal (MST) neurons by using a three-layered autoencoder. Noise effect is taken into account by using the framework of statistical physics. We define a cost function
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::57b9edec31ca63620fa1df7cc3422f9f
https://doi.org/10.1016/s0378-4371(02)01803-4
https://doi.org/10.1016/s0378-4371(02)01803-4
Publikováno v:
Physica A: Statistical Mechanics and its Applications. 317:270-298
We investigate storage capacity and generalization ability for two types of fully connected layered neural networks with non-monotonic transfer functions; random patterns are embedded into the networks by a Hebbian learning rule. One of them is a lay
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::0d38967a7b3ebc5e385d35ef31a9efcd
https://doi.org/10.1016/s0378-4371(02)01319-5
https://doi.org/10.1016/s0378-4371(02)01319-5
Publikováno v:
Physica A: Statistical Mechanics and its Applications. 310:532-546
We investigate storage capacity of two types of fully connected layered neural networks with sparse coding when binary patterns are embedded into the networks by a Hebbian learning rule. One of them is a layered network, in which a transfer function
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d3715facb67c4b249d43278a5c28f2c6
https://doi.org/10.1016/s0378-4371(02)00785-9
https://doi.org/10.1016/s0378-4371(02)00785-9
Autor:
Tsuyoshi Horiguchi, Katsuki Katayama
Publikováno v:
Journal of the Physical Society of Japan. 71:458-465
We investigate an on-line learning of two-layered feed-forward neural networks with randomly diluted connections by using a gradient descent algorithm. We derive coupled first-order differential equations for order parameters which describe learning
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::e9ee5f662164a0efd64c7d478d4b47ee
https://doi.org/10.1143/jpsj.71.458
https://doi.org/10.1143/jpsj.71.458
Autor:
Katsuki Katayama, Tsuyoshi Horiguchi
Publikováno v:
Physica A: Statistical Mechanics and its Applications. 297:532-548
We investigate storage capacity of a fully connected layered neural network with Q(⩾2)-states clock neurons, including Q=∞ (corresponding to oscillatory neurons) and with intra-layer connections, where random Q-values patterns are embedded into t
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::88494258abb340582f4ab9733eadbff5
https://doi.org/10.1016/s0378-4371(01)00264-3
https://doi.org/10.1016/s0378-4371(01)00264-3
Autor:
Katsuki Katayama, Tsuyoshi Horiguchi
Publikováno v:
Journal of the Physical Society of Japan. 70:1300-1314
We investigate storage capacity and retrieval property for a synchronous fully connected neural network with a non-monotonic transfer function which retrieves sequences of patterns, by an analytic method and also by numerical simulations. Because of
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::c32684066ed9f0dc216cc6e786be3e75
https://doi.org/10.1143/jpsj.70.1300
https://doi.org/10.1143/jpsj.70.1300