{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2023,11,17]],"date-time":"2023-11-17T05:55:26Z","timestamp":1700200526972},"reference-count":16,"publisher":"Wiley","issue":"14","license":[{"start":{"date-parts":[[2003,10,14]],"date-time":"2003-10-14T00:00:00Z","timestamp":1066089600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Systems & Computers in Japan"],"published-print":{"date-parts":[[2003,12]]},"abstract":"Abstract<\/jats:title>A complex neural network is obtained from an ordinary network by extending the (real\u2010valued) parameters, such as the weights and the thresholds, to complex values. Applications to problems involving complex numbers, such as communications systems, are expected. This paper presents the following uniqueness theorem. When a complex function is given, the three\u2010layered neural network that approximates the function is uniquely determined by a certain finite group, if it is irreducible. The above finite group specifies the redundancy of the parameters in the complex neural network, but has a structure which is different from that of the real\u2010valued neural network. The order of the finite group is examined, and it is shown that the redundancy of the complex\u2010valued neural network is an exponent multiple of the redundancy of the real\u2010valued neural network. Analysis of the redundancy is important in the theoretical investigation of the basic characteristics of complex\u2010valued neural networks, such as the local minimum property. A sufficient condition is derived for the given three\u2010layered complex\u2010valued neural network to be minimal. The above results are shown, in essence, by extending the approach of Sussmann for real\u2010valued neural networks. \u00a9 2003 Wiley Periodicals, Inc. Syst Comp Jpn, 34(14): 54\u201362, 2003; Published online in Wiley InterScience (www.interscience.wiley.com<\/jats:ext-link>). DOI 10.1002\/scj.10363<\/jats:p>","DOI":"10.1002\/scj.10363","type":"journal-article","created":{"date-parts":[[2003,10,15]],"date-time":"2003-10-15T16:57:02Z","timestamp":1066237022000},"page":"54-62","source":"Crossref","is-referenced-by-count":10,"title":["The uniqueness theorem for complex\u2010valued neural networks and the redundancy of the parameters"],"prefix":"10.1002","volume":"34","author":[{"given":"Tohru","family":"Nitta","sequence":"first","affiliation":[]}],"member":"311","published-online":{"date-parts":[[2003,10,14]]},"reference":[{"key":"e_1_2_1_2_2","first-page":"550","volume-title":"Knowledge\u2010based intelligent information engineering systems and allied technologies","author":"Baba N","year":"2001"},{"key":"e_1_2_1_3_2","doi-asserted-by":"publisher","DOI":"10.1007\/BFb0047683"},{"key":"e_1_2_1_4_2","doi-asserted-by":"publisher","DOI":"10.1109\/78.127967"},{"key":"e_1_2_1_5_2","doi-asserted-by":"publisher","DOI":"10.1016\/S0893-6080(00)00009-5"},{"key":"e_1_2_1_6_2","unstructured":"FukumizuK.Likelihood ratio of unidentifiable models and multilayer neural networks. Res Memo No. 780 The Institute of Statistical Mathematics 2001."},{"key":"e_1_2_1_7_2","first-page":"2058","article-title":"Nonuniqueness of connecting weights and AIC in multi\u2010layer neural networks","volume":"76","author":"Hagiwara K","year":"1993","journal-title":"Trans IEICE"},{"key":"e_1_2_1_8_2","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN.1993.714176"},{"key":"e_1_2_1_9_2","first-page":"1319","article-title":"A complex backpropagation learning","volume":"32","author":"Nitta T","year":"1991","journal-title":"Trans Inf Process Soc"},{"key":"e_1_2_1_10_2","doi-asserted-by":"publisher","DOI":"10.1016\/S0893-6080(97)00036-1"},{"key":"e_1_2_1_11_2","first-page":"612","article-title":"Complex\u2010valued neural networks","volume":"83","author":"Nitta T","year":"2000","journal-title":"Proc IEICE"},{"key":"e_1_2_1_12_2","doi-asserted-by":"publisher","DOI":"10.1023\/A:1026582217675"},{"key":"e_1_2_1_13_2","series-title":"Parallel distributed processing: Explorations in the microstructures of cognition","first-page":"318","author":"Rumelhart DE","year":"1986"},{"key":"e_1_2_1_14_2","doi-asserted-by":"publisher","DOI":"10.1016\/S0893-6080(05)80037-1"},{"key":"e_1_2_1_15_2","first-page":"1363","article-title":"A method to interpret 3D motions using neural networks","volume":"77","author":"Watanabe A","year":"1994","journal-title":"IEICE Trans Fundam"},{"key":"e_1_2_1_16_2","first-page":"356","volume-title":"Advances in neural information processing systems 12","author":"Watanabe S","year":"2000"},{"key":"e_1_2_1_17_2","first-page":"172","volume-title":"Advances in neural information processing systems 5","author":"Zemel RS","year":"1993"}],"container-title":["Systems and Computers in Japan"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1002%2Fscj.10363","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/scj.10363","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,17]],"date-time":"2023-11-17T05:05:38Z","timestamp":1700197538000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/scj.10363"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2003,10,14]]},"references-count":16,"journal-issue":{"issue":"14","published-print":{"date-parts":[[2003,12]]}},"alternative-id":["10.1002\/scj.10363"],"URL":"https:\/\/doi.org\/10.1002\/scj.10363","archive":["Portico"],"relation":{},"ISSN":["0882-1666","1520-684X"],"issn-type":[{"value":"0882-1666","type":"print"},{"value":"1520-684X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2003,10,14]]}}}