{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2023,10,25]],"date-time":"2023-10-25T10:11:38Z","timestamp":1698228698305},"reference-count":9,"publisher":"Wiley","issue":"1","license":[{"start":{"date-parts":[[2007,3,21]],"date-time":"2007-03-21T00:00:00Z","timestamp":1174435200000},"content-version":"vor","delay-in-days":4827,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Systems &amp;amp; Computers in Japan"],"published-print":{"date-parts":[[1994,1]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>The distributed representation\u2010type three\u2010layered perceptron with backpropagation has such problems as the local minimum, long learning time, and ambiguity in internal representation. As a method to cope with those problems, this paper proposes the four\u2010layered perceptron, together with the learning algorithm, where a hidden layer is added, so that each discrete sample point can perfectly be represented by the corresponding output of the upper hidden layer.<\/jats:p><jats:p>First, the learning algorithm of the perceptron is applied successively to the sample points, and the learning is executed so that the input sample points are separated perfectly by the piecewise sets of hyperplanes. In this mechanism, the output matrix of the lower bidder layer output is nonsingular. Consequently, the following four\u2010layered perceptron can be constructed, where the output matrix of the upper hidden layer is an identity matrix, and any discrete value can be produced as the output from the output layer by adjusting the network coefficients. Computational experiments are made for the realization of the three\u2010valued logic function, which is a learning problem on the two\u2010dimensional plane, as well as the pattern recognition problem by the representative sample points. As a result, it is shown that the learning converges in less than 1\/100 computation time, compared to the three\u2010layered perceptron with the backpropagation.<\/jats:p>","DOI":"10.1002\/scj.4690250106","type":"journal-article","created":{"date-parts":[[2007,7,8]],"date-time":"2007-07-08T01:44:09Z","timestamp":1183859049000},"page":"67-77","source":"Crossref","is-referenced-by-count":0,"title":["A perfect separation of discrete sample points by four\u2010layered perceptron with localized representation"],"prefix":"10.1002","volume":"25","author":[{"given":"Takahumi","family":"Oohori","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Naoyuki","family":"Nagao","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kazuhisa","family":"Watanabe","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2007,3,21]]},"reference":[{"issue":"13","key":"e_1_2_1_2_2","first-page":"1811","article-title":"Multilayered neural network","volume":"21","author":"Ae T.","year":"1989","journal-title":"Bit"},{"key":"e_1_2_1_3_2","article-title":"Properties of binary pattern error propagation learning and high speed learning technique","volume":"89","author":"Niki K.","year":"1989","journal-title":"Papers of Technical Group on Neurocomputing, I.E.I.C.E., Japan"},{"key":"e_1_2_1_4_2","article-title":"Recognition performance for character pattern with superposed noise by multilayered neural network","volume":"90","author":"Ohguchi Y.","year":"1990","journal-title":"Papers of Technical Group on Neurocomputing, I.E.I.C.E., Japan"},{"issue":"12","key":"e_1_2_1_5_2","first-page":"101","article-title":"On\u2010line handwritten kanji character recognition by improved perceptron considering recognition order","volume":"111","author":"Oohori T.","year":"1990","journal-title":"I.E.E., Japan, D"},{"key":"e_1_2_1_6_2","volume-title":"Perceptron","author":"Minsky M.","year":"1969"},{"key":"e_1_2_1_7_2","doi-asserted-by":"publisher","DOI":"10.1038\/323533a0"},{"issue":"10","key":"e_1_2_1_8_2","first-page":"1465","article-title":"Function approximation by three\u2010layered perceptron with localized representation","volume":"74","author":"Watanabe K.","year":"1991","journal-title":"Trans. (D\u2010II) I.E.I.C.E., Japan"},{"key":"e_1_2_1_9_2","unstructured":"K.Funabashi.Capability of neural network. Papers of Technical Group on Medical and Biological Engineering I.E.I.C.E. Japan MBE88\u201352 (1988)."},{"issue":"11","key":"e_1_2_1_10_2","first-page":"1872","article-title":"Determination of number of hidden layer units in three\u2010layered neural net by information measure","volume":"73","author":"Kurita T.","year":"1990","journal-title":"Trans. (D\u2010II) I.E.I.C.E., Japan"}],"container-title":["Systems and Computers in Japan"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1002%2Fscj.4690250106","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/scj.4690250106","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,10,24]],"date-time":"2023-10-24T05:56:29Z","timestamp":1698126989000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/scj.4690250106"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[1994,1]]},"references-count":9,"journal-issue":{"issue":"1","published-print":{"date-parts":[[1994,1]]}},"alternative-id":["10.1002\/scj.4690250106"],"URL":"https:\/\/doi.org\/10.1002\/scj.4690250106","archive":["Portico"],"relation":{},"ISSN":["0882-1666","1520-684X"],"issn-type":[{"value":"0882-1666","type":"print"},{"value":"1520-684X","type":"electronic"}],"subject":[],"published":{"date-parts":[[1994,1]]}}}