<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "JATS-journalpublishing1-3.dtd">
<article article-type="research-article" dtd-version="1.3" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xml:lang="ru"><front><journal-meta><journal-id journal-id-type="publisher-id">inform</journal-id><journal-title-group><journal-title xml:lang="ru">Информатика</journal-title><trans-title-group xml:lang="en"><trans-title>Informatics</trans-title></trans-title-group></journal-title-group><issn pub-type="ppub">1816-0301</issn><issn pub-type="epub">2617-6963</issn><publisher><publisher-name>UIIP NASB</publisher-name></publisher></journal-meta><article-meta><article-id custom-type="elpub" pub-id-type="custom">inform-738</article-id><article-categories><subj-group subj-group-type="heading"><subject>Research Article</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="ru"><subject>ИНТЕЛЛЕКТУАЛЬНЫЕ СИСТЕМЫ</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="en"><subject>INTELLIGENT SYSTEMS</subject></subj-group></article-categories><title-group><article-title>ТЕСТОВАЯ ЗАДАЧА ДЛЯ ВИЗУАЛИЗАЦИИ ПРОЦЕССА ОБУЧЕНИЯ НЕЙРОННОЙ СЕТИ</article-title><trans-title-group xml:lang="en"><trans-title></trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Сарапас</surname><given-names>В. В.</given-names></name></name-alternatives><xref ref-type="aff" rid="aff-1"/></contrib></contrib-group><aff xml:lang="ru" id="aff-1"><institution>Объединенный институт проблем информатики НАН Беларуси</institution><country>Belarus</country></aff><pub-date pub-type="collection"><year>2005</year></pub-date><pub-date pub-type="epub"><day>09</day><month>01</month><year>2019</year></pub-date><volume>0</volume><issue>1(5)</issue><fpage>58</fpage><lpage>67</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Сарапас В.В., 2019</copyright-statement><copyright-year>2019</copyright-year><copyright-holder xml:lang="ru">Сарапас В.В.</copyright-holder><copyright-holder xml:lang="en">Сарапас В.В.</copyright-holder><license xml:lang="ru" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>Данная работа распространяется под лицензией Creative Commons Attribution 4.0.</license-p></license><license xml:lang="en" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>This work is licensed under a Creative Commons Attribution 4.0 License.</license-p></license></permissions><self-uri xlink:href="https://inf.grid.by/jour/article/view/738">https://inf.grid.by/jour/article/view/738</self-uri><abstract><p>При разработке и реализации любого нейросетевого алгоритма возникает задача тестирования его работоспособности. В таких случаях используются различные тестовые задачи. При этом, как и в других областях научной деятельности, исследователю приходится использовать методы визуализации данных с целью более эффективного их анализа. Однако большинство тестовых задач не позволяют наглядно и объективно судить о ходе поиска решения в пространстве функции ошибки. В статье дан краткий обзор методов, применяемых для визуализации процесса обучения нейронной сети, а также описан подход к тестированию алгоритмов обучения нейронных сетей, основанный на использовании тестовых задач, обладающих известной функцией ошибки. В качестве иллюстрации к данному подходу  предложена простая тестовая задача для нейронной сети минимального размера, которая позволяет наглядно визуализировать процесс обучения и объективно судить о его ходе.</p></abstract></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">Everitt B.S. Graphical techniques for multivariate data. – London: Heinemann educational books, 1978.</mixed-citation><mixed-citation xml:lang="en">Everitt B.S. Graphical techniques for multivariate data. – London: Heinemann educational books, 1978.</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">Manly B.F.J. Multivariate statistical methods: A primer. – London: Chapman and Hall, 1994.</mixed-citation><mixed-citation xml:lang="en">Manly B.F.J. Multivariate statistical methods: A primer. – London: Chapman and Hall, 1994.</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">Phillips S. The effect of representation on error surface // Fourth Australian conference on neural networks (ACNN’93). – Australia: University of Sydney, 1993. – P. 86–89.</mixed-citation><mixed-citation xml:lang="en">Phillips S. The effect of representation on error surface // Fourth Australian conference on neural networks (ACNN’93). – Australia: University of Sydney, 1993. – P. 86–89.</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Hush D.R., Horne B., Salas J. M. Error surfaces for multi-layer perceptrons // IEEE Transactions on systems, man, and cybernetics. – V. 22. – № 5. – 1992. – P. 1152–1161.</mixed-citation><mixed-citation xml:lang="en">Hush D.R., Horne B., Salas J. M. Error surfaces for multi-layer perceptrons // IEEE Transactions on systems, man, and cybernetics. – V. 22. – № 5. – 1992. – P. 1152–1161.</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">Hamey L.G.C. The structure of neural network error surfaces // Proc. Sixth Australian сonference on neural networks. – University of Sydney, 1995. – P.197–200.</mixed-citation><mixed-citation xml:lang="en">Hamey L.G.C. The structure of neural network error surfaces // Proc. Sixth Australian сonference on neural networks. – University of Sydney, 1995. – P.197–200.</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">Androulakis G.S., Magoulas G.D., Vrahatis M. N. Geometry of learning: Visualizing the performance of neural network supervised training methods // Nonlinear analysis, theory, methods &amp; applications. – V. 30. – № 7. – 1997. – P. 4359–4544.</mixed-citation><mixed-citation xml:lang="en">Androulakis G.S., Magoulas G.D., Vrahatis M. N. Geometry of learning: Visualizing the performance of neural network supervised training methods // Nonlinear analysis, theory, methods &amp; applications. – V. 30. – № 7. – 1997. – P. 4359–4544.</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">Jolliffe T. Principal component analysis. – New-York: Springer-Verlag, 1986.</mixed-citation><mixed-citation xml:lang="en">Jolliffe T. Principal component analysis. – New-York: Springer-Verlag, 1986.</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">Wejchert J. and Tesauro G. Neural network visualization // Advances in neural information processing systems. – V. 2. – 1990. – P. 465–472.</mixed-citation><mixed-citation xml:lang="en">Wejchert J. and Tesauro G. Neural network visualization // Advances in neural information processing systems. – V. 2. – 1990. – P. 465–472.</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">Nilsson N.J. Learning machines. – New-York: McGraw-Hill, 1965.</mixed-citation><mixed-citation xml:lang="en">Nilsson N.J. Learning machines. – New-York: McGraw-Hill, 1965.</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Hinton G.E., McClelland J.L. and Rumelhart D.E. Distributed representations // Parallel distributed processing. – V.1. – Chapter 3. – 1986. – P. 77–109.</mixed-citation><mixed-citation xml:lang="en">Hinton G.E., McClelland J.L. and Rumelhart D.E. Distributed representations // Parallel distributed processing. – V.1. – Chapter 3. – 1986. – P. 77–109.</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">UCI Machine learning repository // http://www.ics.uci.edu/~mlearn/MLRepository.html</mixed-citation><mixed-citation xml:lang="en">UCI Machine learning repository // http://www.ics.uci.edu/~mlearn/MLRepository.html</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">UCI Knowledge discovery in databases // http://kdd.ics.uci.edu/</mixed-citation><mixed-citation xml:lang="en">UCI Knowledge discovery in databases // http://kdd.ics.uci.edu/</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">Delve: Data for Evaluating Learning in Valid Experiments // http://www.cs.toronto.edu/~delve/</mixed-citation><mixed-citation xml:lang="en">Delve: Data for Evaluating Learning in Valid Experiments // http://www.cs.toronto.edu/~delve/</mixed-citation></citation-alternatives></ref><ref id="cit14"><label>14</label><citation-alternatives><mixed-citation xml:lang="ru">Bilkent university function approximation repository // http://funapp.cs.bilkent.edu.tr/funapp/</mixed-citation><mixed-citation xml:lang="en">Bilkent university function approximation repository // http://funapp.cs.bilkent.edu.tr/funapp/</mixed-citation></citation-alternatives></ref><ref id="cit15"><label>15</label><citation-alternatives><mixed-citation xml:lang="ru">CEDAR: Database of handwritten cities, states, ZIP codes, digits, and alphabetic characters // http://www.cedar.buffalo.edu/Databases/</mixed-citation><mixed-citation xml:lang="en">CEDAR: Database of handwritten cities, states, ZIP codes, digits, and alphabetic characters // http://www.cedar.buffalo.edu/Databases/</mixed-citation></citation-alternatives></ref><ref id="cit16"><label>16</label><citation-alternatives><mixed-citation xml:lang="ru">Otago speech corpus // http://divcom.otago.ac.nz/infosci/kel/software/RICBIS/hyspeech_main.html</mixed-citation><mixed-citation xml:lang="en">Otago speech corpus // http://divcom.otago.ac.nz/infosci/kel/software/RICBIS/hyspeech_main.html</mixed-citation></citation-alternatives></ref><ref id="cit17"><label>17</label><citation-alternatives><mixed-citation xml:lang="ru">Lisboa P.J.G., Perantonis S.J. Complete solution of the local minima in the XOR problem // Network. – V. 2. –1991. – P. 119–124.</mixed-citation><mixed-citation xml:lang="en">Lisboa P.J.G., Perantonis S.J. Complete solution of the local minima in the XOR problem // Network. – V. 2. –1991. – P. 119–124.</mixed-citation></citation-alternatives></ref><ref id="cit18"><label>18</label><citation-alternatives><mixed-citation xml:lang="ru">Hamney L.G.C. XOR has no local minima: a case study in neural network error surface analysis // Neural Networks. – V. 11 – № 4. – 1998. – P. 669–682.</mixed-citation><mixed-citation xml:lang="en">Hamney L.G.C. XOR has no local minima: a case study in neural network error surface analysis // Neural Networks. – V. 11 – № 4. – 1998. – P. 669–682.</mixed-citation></citation-alternatives></ref><ref id="cit19"><label>19</label><citation-alternatives><mixed-citation xml:lang="ru">Sprinkhuizen-Kuyper I.G., Boers E.J.W. The error surface of the 2-2-1 XOR network: the finite stationary points // Neural Networks. – V. 11 – № 4. – 1998. – P. 683–690.</mixed-citation><mixed-citation xml:lang="en">Sprinkhuizen-Kuyper I.G., Boers E.J.W. The error surface of the 2-2-1 XOR network: the finite stationary points // Neural Networks. – V. 11 – № 4. – 1998. – P. 683–690.</mixed-citation></citation-alternatives></ref><ref id="cit20"><label>20</label><citation-alternatives><mixed-citation xml:lang="ru">Auer P., Herbster M. and Warmuth M.K. Exponentially many local minima for single neurons // Technical Report UCSC-CRL-96-1. – Univ. of Calif. computer research lab., 1996.</mixed-citation><mixed-citation xml:lang="en">Auer P., Herbster M. and Warmuth M.K. Exponentially many local minima for single neurons // Technical Report UCSC-CRL-96-1. – Univ. of Calif. computer research lab., 1996.</mixed-citation></citation-alternatives></ref></ref-list><fn-group><fn fn-type="conflict"><p>The authors declare that there are no conflicts of interest present.</p></fn></fn-group></back></article>
