Article

Article title USE OF TECHNOLOGY TRANSFER LEARNING FOR INTELLECTUAL FORMATION OF STARTING POPULATIONS FOR SOLUTION OF TASKS OF TRANSPORT TYPE
Authors Yu. O. Chernyshev, N. N. Ventsov, I. S. Pshenichniy
Section SECTION II. ARTIFICIAL INTELLIGENCE AND FUZZY SYSTEMS
Month, Year 04, 2018 @en
Index UDC 681.3
DOI
Abstract The paper presents a method of forming new terms describing fuzzy solutions of optimization problems of transport type. A distinctive feature of the proposed method is the use of Transfer Learning technology and a parametric approach to the construction of membership functions. The technology Transfer Learning allows you to transfer data between context-related tasks. The context-related task that is being solved at a given time is the target, its peculiarity is the presence of uncertainties due to incomplete or contradictory description of the modeled domain. The initial one is understood to mean a problem with a large number, in comparison with the target, of known (formalized) components such as objective function, constraint system, input data, etc. It is understood that the fragments of the source and target tasks are in some contextual relationship. Mathematical models of modern optimization problems contain both clear and fuzzy components, so the problem of transferring fuzzy parameters becomes topical. A significant drawback of the technology Transfer Learning is the use of the same type of transformation when transferring the data of the original task to the target. Using the parametric approach allows you to adapt the data transferred to the intermediate solution of the target task. Adaptation consists in organizing the movement of the individuals of the starting population in the fuzzy search space. The peculiarity of this movement is the possibility of using the results, borrowed with the Transfer Learning technology, not only as arguments of the function by which the direction of motion is determined, but also for determining the transition coefficients specifying the features of the motion path determination functions. The adaptation strategy can be defined using known intellectual approaches, for example, random search algorithms. The stronger the contextual interconnectedness is the target and solved problem, the more effective the use of the proposed approach becomes.

Download PDF

Keywords Fuzzy systems; transfer of knowledge; adaptation; intellectual methods; context.
References 1. Bova V.V., Gladkov L.A., Kravchenko YU.A., Kureychik V.V., Kureychik V.M., Nuzhnov E.V., Rogozov YU.I., Sviridov A.S., Sorokoletov P.V., Shcheglov S.N. Tekhnologii intellektual'nogo analiza i izvlecheniya dannykh na osnove principov evolyucionnogo modelirovaniya [Technologies of data mining and extraction based on the principles of evolutionary modeling]. Taganrog: TTI YUFU, 2009, 124 p.
2. Ross D. Strukturnyy analiz (SA): yazyk dlya peredachi ponimaniya. Trebovaniya i speci-fikacii v razrabotke program [Structural analysis (SA): language to convey understanding. Requirements and specifications in software development]. Moscow: Mir, 1984, 344 p.
3. Sviridov A.S. Metod postroeniya matematicheskoy modeli informacionnykh potokov predpriyatiya [Method of constructing a mathematical model of information flows of the enterprise], Izvestiya TRTU [Izvestiya TSURE], 2004, No. 3 (38), pp. 152-156.
4. Chernyshev Yu.O., Basova A.V., Poluyan A.Yu. Reshenie zadach transportnogo tipa geneticheskimi algoritmami [Solution of transport type problems by genetic algorithms]. Rostov-on-Don: Izd-vo YUFU GOU, 2008, 87 p.
5. Chernyshev Yu.O., Basova A.V., Panasenko P.A., Polyakov V.I. Ispol'zovanie metodov modelirovaniya evolyucii dlya optimizacii dokumentooborota na predpriyatii [Use of methods of evolution modeling for optimization of document flow at the enterprise], Vestnik informacionnykh tekhnologiy, mekhaniki i optiki [Bulletin of information technologies, mechanics and optics], 2013, No. 1 (83), pp. 135-140.
6. Kureнchik V.M., Lebedev B.K., Lebedev O.B. Poiskovaya adaptaciya: teoriya i praktika: monografiya [Search adaptation: theory and practice: monograph]. Moscow: Fizmatlit, 2006, 272 p.
7. Lebedev B.K., Lebedev O.B., Chernyshev Yu.O. Osnovnye zadachi sinteza topologii SBIS: monografiya [The main tasks of synthesis of VLSI topology: monograph]. Rostov-on-Don: RGASHM, 2006, 92 p.
8. Zade L.A. Fuzzy sets, Information and Control, 1965, Vol. 8, 338 p.
9. Polkovnikova N.A., Kureychik V.M. Razrabotka modeli ekspertnoy sistemy na osnove nechyotkoy logiki [Development of a model of an expert system based on fuzzy logic], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2014, No. 1 (150), pp. 83-92.
10. Bettini C., Brdiczka O., Henricksen K., Indulska J., Nicklas D., Ranganathan A., Riboni D.A survey of context modelling and reasoning techniques, Pervasive and Mobile Computing, 2010, .No. 6 (2), pp. 161-180.
11. Shell J, Coupland S. Fuzzy Transfer Learning: Methodology and Application, Preprint submitted to Information Sciences, May 23, 2014, 27 p.
12. Vencov N.N., Dolmatov A.A., Chernyshev Yu.O. Evolyucionnyy algoritm resheniya nechetko sformulirovannoy transportnoy zadachi [Evolutionary algorithm for solving fuzzy transport problem], IS-IT`17: Tr. Mezhdunar. kongr. po intellekt. sistemam i inform. tekhnologiyam, p. Divnomorskoe, 2-9 sentyabrya [IS-IT`17: Proceedings of The international Congress on intelligent systems and information technologies, p. Divnomorskoe, September 2-9]. Tananrog: YuFU, 2017, Vol. 1, pp. 13-19.
13. Borisov A.N., Krumberg O.A., Fedorov I.P. Prinyatie resheniy na osnove nechetkikh modeley: Primery ispol'zovaniya [Decision making based on fuzzy models: use Cases]. Riga: Zinatne, 1990, 184 p.
14. Mendel J.M. Fuzzy-logic systems for engineering, A. tutorial. Proceedings of IEEE, 1995, Vol. 83 (3), pp. 345-377.
15. Shell J, Coupland S. Fuzzy Transfer Learning: Methodology and Application, Preprint submitted to Information Sciences May 23, 2014, 27 p.
16. Nazim M, Hashim M, Xu J. Multi Objective Optimization of Production-Distribution Problem under Fuzzy Random Environment, Global J. Technol. Optim., 2014, No. 5: 161. Doi: 10.4172/2229-8711.1000161.
17. Enrique López González, Miguel A. Rodríguez Fernández, Cristina Mendaña-Cuervo, Raquel Flórez López. The distribution problem in management accounting with genetic algorithm and fuzzy sets, Proceedings of the EUSFLAT-ESTYLF Joint Conference, Palmade Mallorca, Spain, September 22-25, 1999.
18. Cristina Mendaña-Cuervo, Enrique López-González, Begoña González-Pérez. A Model of Genetic Fuzzy System for the Design of New Products, Proceedings of the Joint 4th Conference of the European Society for Fuzzy Logic and Technology and the 11th Rencontres Francophones sur la LogiqueFloue et ses Applications, Barcelona, Spain, September 7-9, 2005.
19. Chernyshev Yu.O., Vencov N.N., Dolmatov A.A. Sposob perenosa dannyh mezhdu kontekstno svyazannymi zadachami na osnove PSO-metoda [Method of data transfer between context-related tasks based on PSO-method], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2017, No. 7 (192), pp. 154-162.
20. Nechetkie mnozhestva v modelyakh upravleniya i iskusstvennogo intellekta [Fuzzy sets in control and artificial intelligence], ed. by D.A. Pospelova. Moscow: Nauka, 1986, 321 p.
21. Kureychik V.M., Safronenkova I.B. Intellektual'naya klassifikaciya v usloviyakh shuma [Intelligent classification in noise conditions], Tr. konferencii «Iskusstvennyy intellekt: problemy i puti resheniya», 14-15 marta 2018. FGAU «Kongressno-vystavochnyy centr «Patriot» [Proceedings of the conference "Artificial intelligence: problems and solutions", 14-15 March 2018. FSAU "Congress and exhibition center "Patriot"], pp. 17-23.
22. Zheltov S.Yu., Fedunov B.E. Raspredelennyy bortovoy iskusstvennyy intellekt podderzhki processa resheniya takticheskikh zadach ekipazhami letatel'nykh apparatov [Distributed on-Board artificial intelligence to support the process of solving tactical problems by crews of aircraft], Tr. konferencii «Iskusstvennyy intellekt: problemy i puti resheniya», 14-15 marta 2018. FGAU «Kongressno-vystavochnyy centr «Patriot» [Proceedings of the conference "Artificial intelligence: problems and solutions", 14-15 March 2018. FGAU "Congress and exhibition center "Patriot"], pp. 17-23.

Comments are closed.