{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2023,10,24]],"date-time":"2023-10-24T11:41:55Z","timestamp":1698147715191},"reference-count":38,"publisher":"Wiley","issue":"3","license":[{"start":{"date-parts":[[2006,10,25]],"date-time":"2006-10-25T00:00:00Z","timestamp":1161734400000},"content-version":"vor","delay-in-days":4560,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Concurrency: Pract. Exper."],"published-print":{"date-parts":[[1994,5]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>We present two strategies for the simulation of massive neural networks on message\u2010passing MIMD machines. In the first strategy all interconnections between neurons are stored explicitly in interconnection matrices. During simulation, every processor is responsible for certain submatrices of these interconnection matrices. The fact that message\u2010passing MIMD processors do not provide virtual memory seriously limits the size of the networks that can be simulated, since interconnection matrices require huge amounts of memory.<\/jats:p><jats:p>An alternative strategy is not to store the connections explicitly, but generate the interconnections as they are needed. This circumvents memory limitations, but because interconnections need to be generated multiple times, it is inherently slower than the first implementation.<\/jats:p><jats:p>This yields the connections dilemma: the choice between fast simulation of small networks as against slower simulation of massive networks.<\/jats:p><jats:p>We present, analyze and bench\u2010mark parallel implementations for both strategies. An efficient connection\u2010look\u2010up algorithm, which can be used for any network with static interconnections, ensures that simulation times for the second strategy are only marginally longer than for the first strategy. We show that for our users the connections dilemma is no longer a dilemma: by means of our look\u2010up algorithm the simulation of massive networks becomes possible; furthermore the time to design and construct a network, prior to simulation, is considerably shorter than it is for the matrix version, and in addition this time is independent of network size.<\/jats:p><jats:p>Although we have implemented both strategies on a parallel computer, the algorithms presented here can be used on any machine with memory limitations, such as personal computers.<\/jats:p>","DOI":"10.1002\/cpe.4330060302","type":"journal-article","created":{"date-parts":[[2006,11,18]],"date-time":"2006-11-18T06:53:43Z","timestamp":1163832823000},"page":"153-191","source":"Crossref","is-referenced-by-count":0,"title":["Massive MIMD neural network simulations: The connections dilemma"],"prefix":"10.1002","volume":"6","author":[{"given":"Tom","family":"Tollenaere","sequence":"first","affiliation":[]},{"given":"J. Miguel","family":"Saraiva","sequence":"additional","affiliation":[]},{"given":"Marc M.","family":"Van Hulle","sequence":"additional","affiliation":[]}],"member":"311","published-online":{"date-parts":[[2006,10,25]]},"reference":[{"key":"e_1_2_1_2_2","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.79.8.2554"},{"key":"e_1_2_1_3_2","first-page":"397","volume-title":"Methods in Neuronal Modeling","author":"Nelson M. E.","year":"1989"},{"key":"e_1_2_1_4_2","doi-asserted-by":"publisher","DOI":"10.1016\/0743-7315(89)90063-4"},{"key":"e_1_2_1_5_2","volume-title":"Proc. Third Conf. on Hypercube Concurrent Computers and Applications","author":"Bower J. M.","year":"1988"},{"key":"e_1_2_1_6_2","doi-asserted-by":"publisher","DOI":"10.1016\/0743-7315(92)90068-X"},{"key":"e_1_2_1_7_2","doi-asserted-by":"publisher","DOI":"10.1126\/science.3175666"},{"key":"e_1_2_1_8_2","unstructured":"J. M.Saraiva \u2018Simulation of neural networks on transputer networks\u2019 Thesis Computer Engineering Informatics Department Universidade do Minho Braga Portugal 1991."},{"key":"e_1_2_1_9_2","doi-asserted-by":"crossref","unstructured":"Tollenaere T.andOrban G. A. \u2018Simulating modular neural networks on message\u2010passing multiprocessors\u2019 Parallel Comput. 361\u2013379(1991).","DOI":"10.1016\/S0167-8191(05)80142-5"},{"key":"e_1_2_1_10_2","doi-asserted-by":"publisher","DOI":"10.1016\/0743-7315(92)90069-Y"},{"key":"e_1_2_1_11_2","doi-asserted-by":"publisher","DOI":"10.1017\/S0033583500004492"},{"key":"e_1_2_1_12_2","volume-title":"Self Organization and Associative Memory","author":"Kohonen T.","year":"1984"},{"key":"e_1_2_1_13_2","doi-asserted-by":"publisher","DOI":"10.1109\/MASSP.1987.1165576"},{"key":"e_1_2_1_14_2","doi-asserted-by":"crossref","DOI":"10.7551\/mitpress\/5236.001.0001","volume-title":"Parallel and Distributed Processing. Explorations into the Microstructure of Cognition","author":"Rumelhart D. E.","year":"1986"},{"key":"e_1_2_1_15_2","doi-asserted-by":"publisher","DOI":"10.1002\/spe.4380201002"},{"key":"e_1_2_1_16_2","volume-title":"Proc. Computer Vision Workshop","author":"Forrest B.","year":"1987"},{"key":"e_1_2_1_17_2","first-page":"254","volume-title":"Neural Networks from Models to Applications","author":"Van Hulle M. M.","year":"1989"},{"key":"e_1_2_1_18_2","doi-asserted-by":"publisher","DOI":"10.1016\/S0893-6080(05)80070-X"},{"key":"e_1_2_1_19_2","first-page":"256","article-title":"Entropy driven networks for sensorial representation: a proposal","volume":"6","author":"Van Hulle M. M.","year":"1989","journal-title":"J. Parallel Distrib. Comput."},{"key":"e_1_2_1_20_2","first-page":"107","volume-title":"Transputing '91","author":"Tollenaere T.","year":"1991"},{"key":"e_1_2_1_21_2","volume-title":"Transputer Reference Manual","author":"INMOS Ltd.","year":"1988"},{"key":"e_1_2_1_22_2","volume-title":"CSTools for SunOS","author":"Meiko Scientific","year":"1991"},{"key":"e_1_2_1_23_2","unstructured":"T.Tollenaere. \u2018The simulation of entropy driven artificial neural networks on message\u2010passing multiprocessors\u2019 PhD thesis Department of Computer Science K. U. Leuven 1993."},{"key":"e_1_2_1_24_2","volume-title":"Solving Problems on Concurrent Processors \u2010 Vol. 1: General Techniques and Regular Problems","author":"Fox G.","year":"1988"},{"key":"e_1_2_1_25_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-94-009-0643-3_7"},{"key":"e_1_2_1_26_2","first-page":"173","volume-title":"Developments Using occam","author":"Richards G.","year":"1988"},{"key":"e_1_2_1_27_2","first-page":"721","article-title":"Stochastic reiaxation. Gibbs distributions, and the Bayesian restoration of images","volume":"5","author":"Geman S.","year":"1984","journal-title":"IEEE Trans."},{"key":"e_1_2_1_28_2","doi-asserted-by":"crossref","first-page":"141","DOI":"10.1007\/BF00339943","article-title":"Neural computations of decisions in optimization problems","volume":"52","author":"Hopfield J. J.","year":"1985","journal-title":"Biological Cybernetics"},{"key":"e_1_2_1_29_2","doi-asserted-by":"publisher","DOI":"10.1016\/0262-8856(86)90056-9"},{"key":"e_1_2_1_30_2","doi-asserted-by":"publisher","DOI":"10.1364\/JOSAA.2.001160"},{"key":"e_1_2_1_31_2","doi-asserted-by":"publisher","DOI":"10.1098\/rspb.1982.0078"},{"key":"e_1_2_1_32_2","volume-title":"Methods in Neuronal Modeling. From Synapses to Networks","author":"Shamma S.","year":"1989"},{"key":"e_1_2_1_33_2","doi-asserted-by":"publisher","DOI":"10.1016\/0166-2236(87)90126-3"},{"key":"e_1_2_1_34_2","doi-asserted-by":"publisher","DOI":"10.1016\/0165-6074(88)90120-2"},{"key":"e_1_2_1_35_2","doi-asserted-by":"crossref","first-page":"3188","DOI":"10.1523\/JNEUROSCI.09-09-03188.1989","article-title":"Integration of distributed cortical systems by reentry: a computer simulation of interactive functionally segregated visual areas","volume":"9","author":"Finkel L. H.","year":"1989","journal-title":"J. Neurosc."},{"key":"e_1_2_1_36_2","first-page":"408","volume-title":"Models of the Visual Cortex","author":"Koch C.","year":"1985"},{"key":"e_1_2_1_37_2","volume-title":"Studies of Brain Functions, 11","author":"Orban G. A.","year":"1984"},{"key":"e_1_2_1_38_2","volume-title":"Advanced Methods of Physiological System Modeling","author":"Palmer L.","year":"1987"},{"key":"e_1_2_1_39_2","first-page":"37","volume-title":"Transputer Systems \u2010 Ongoing Research","author":"Tollenaere T.","year":"1992"}],"container-title":["Concurrency: Practice and Experience"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1002%2Fcpe.4330060302","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/cpe.4330060302","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,10,24]],"date-time":"2023-10-24T00:22:19Z","timestamp":1698106939000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/cpe.4330060302"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[1994,5]]},"references-count":38,"journal-issue":{"issue":"3","published-print":{"date-parts":[[1994,5]]}},"alternative-id":["10.1002\/cpe.4330060302"],"URL":"https:\/\/doi.org\/10.1002\/cpe.4330060302","archive":["Portico"],"relation":{},"ISSN":["1040-3108","1096-9128"],"issn-type":[{"value":"1040-3108","type":"print"},{"value":"1096-9128","type":"electronic"}],"subject":[],"published":{"date-parts":[[1994,5]]}}}