DOI: 10.17587/prin.14.69-76

System of Operations on Associative Heterarchical Memory

R. V. Dushkin, Chief Science and Technology Officer, drv@aiagency.ru, V. A. Lelecova, Analyst, lv@aiagency.ru, K. Yu. Eidemiller, Academic Chief, Artificial Intelligence Agency, Moscow, 127473, Russian Federation

Corresponding author: Vasilisa A. Lelecova, Analyst, Artificial Intelligence Agency, Moscow, 127473, Russian Federation, E-mail: lv@aiagency.ru

Text processing in natural language remains an important task for the field of development of artificial intelligence methods and tools. Since the twentieth century, artificial intelligence methods have been divided into two paradigms — top-down and bottom-up. The methods of the ascending paradigm are difficult to interpret in the form of the output of natural language, and the methods of the descending paradigm are difficult to actualize information. Taking into account the authors approach to the construction of artificial intelligence agents, the processing of natural language must be performed on two levels: on the lower level, using methods of the bottom-up paradigm, and on the upper level, using symbolic methods of the top-down paradigm. The authors of the article have already introduced a new mathematical formalism based on the notion of a hypergraph — associative heterarchical memory (AH-memory). Such memory should simplify the process of natural language processing with new technologies. Earlier the authors group has thoroughly analyzed the problem of symbol binding in the application to АН-memory and its structure. In the first paper, abstract symbol binding was performed using multi-serial integration, eventually converting the primary symbols received by the program into integrated abstract symbols. The second paper provided a comprehensive description of the AH-memory in the form of formulas, explanations of them, and their corresponding diagrams. Although there are many possible modules to use, the developer working with AH-memory should choose those parts of AH-memory which are required for successful and efficient functioning of the AI agent. The article will be of interest to developers of artificial intelligence methods and tools, mathematicians and specialists in natural language processing.

Keywords: associative heterarchical memory, natural language processing, hypergraph, mathematical formalism, artificial intelligence, symbolic method, abstract symbols, operation system, AI agent

pp. 69–76

For citation:
Dushkin R. V., Lelecova V. A., Eindemiller K. Yu. System of Operations on Associative Heterarchical Memory, Programmnaya Ingeneria, 2023, vol. 14, no. 2, pp. 69—76. DOI: 10.17587/prin.14.69-76 (in Russian).

References:

1. Raina V., Krishnamurthy S. Natural Language Processing, Building an Effective Data Science Practice. Apress, Berkeley, CA. 2022, pp. 63—73. DOI: 10.1007/978-1-4842-7419-4_6.
2. Dushkin R. V. Overview of Artificial Intelligence Approaches and Methods, Radioelectronnye techlonogii, 2018, no. 3, pp. 85—89 (in Russian).
3. Dushkin R. V. Artificial Intelligence, Moscow, DMK-Press, 2019, 280 p. (in Russian).
4. Zadeh L. A. From computing with numbers to computing with words — From manipulation of measurements to manipulation of perceptions, Int. J. Appl. Math. Comput.Sci., 2001, vol. 12, no. 3, pp. 307—324.
5. Dushkin R. V. Development of adaptive learning methods using intelligent agents, Iskusstvenniy intellect i prinyatie resheniy, 2019, no. 1, pp. 87—96 (in Russian).
6. Xu Yan, Yining Wang, Tianren Liu et al. An end-to-end system to identify temporalrelation in discharge summaries: 2012 i2b-2challenge, Journal of the American Medical Informatics Association, 2012, vol. 20, no. 5, pp. 849—858. DOI: 10.1136/amiajnl-2012-001607.
7. Sohrab M. G., Khoa D., Makoto M. et al. BENNERD: A neural namedentity linkingsystem for COVID-19, Proceedings of the 2020 Conference on Empirical Methods in Natural Language Pro­cessing: System Demonstrations, 2020, pp. 182—188. DOI: 10.18653/ v1/2020.emnlp-demos.24.
8. Kuniyoshi Fusataka, Jun Ozawa, Mikiya Fujii et al. Graph representation forsynthesis process extraction frominorganic material literature, IEICETechnical Report, 2019, vol. 119 (212), pp. 7—12.
9. Yoshinobu K., Miwa M., Cohen K. B. et al. U-Compare: A modular NLP workflow construction and evaluationsystem, IBM Journal of Research and Development, 2011, vol. 55, no. 3, pp. 11. DOI: 10.1147/JRD.2011.2105691.
10. Dushkin R. V., Lelekova V. A., Stepankov V. Y., Fadeeva S. The structure of associative heterarchical memory, SSRN Electronic Journal, 2022. DOI: 10.2139/ssrn.4196439/
11. Zhu L., Gao W. Hypergraph Ontology Sparse Vector Representation and Its Application to Ontology Learning, Data Mining and Big Data. DMBD 2021, Communications in Computer and Information Science / Y. Tan, Y. Shi, A. Zomaya, H. Yan, J. Cai (eds), Springer, Singapore, 2021, vol 1454. DOI: 10.1007/978-981-16-7502-7_2.
12. Harnad S. The Symbol Grounding Problem, Physica, 1990. D 42, pp. 335—346, available at: https://bit.ly/3z5mHcl (date of access 29.12.2021).
13. Cormen T. Lejzerson Ch., Rivest R., Shtain K. Introduction to algorithms. 2nd ed. Moscow, Wiliams, 2006, 1296 p. (in Russian).
14. Jezequel J.-M. Domain Specific Languages: From Craft to Engineering, iiWAS'14: Proceedings of the 16th International Conference on Information Integration and Web-based Applications & Services, 2014, 2 p. DOI: 10.1145/2684200.2684370.
15. Flemmer R. C. A scheme for an embodied artificial intelligence, 2009 4th International Conference on Autonomous Robots and Agents, 2009, pp. 1—9. DOI: 10.1109/ICARA.2000.4804031.
16. Makondo Ndivhuwo. Accelerating robot learning of motor skills with knowledge transfer, Thesis for: Doctor of Philosophy in Computational Intelligence and Systems Science, 2018.DOI: 10.13140/RG.2.2.26694.32329.