Computational Intelligence

Major: Systems and Methods of Decision Making
Code of subject: 7.124.01.O.002
Credits: 6.00
Department: Information Systems and Networks
Lecturer: Doctor of Sciences., Professor Peleshchak Roman Mykhaylovych
Semester: 1 семестр
Mode of study: денна
Мета вивчення дисципліни: The purpose of the discipline is to study the technologies of models, methods and software tools for solving informal, creative tasks in various spheres of activity using the apparatus and logic of human thinking in decision-making, classification, pattern recognition, etc.
Завдання: The study of an educational discipline involves the formation of competencies in students of education: general competences: The ability to solve complex specialized tasks and practical problems characterized by the complexity and uncertainty of conditions and requirements in various fields, which involves conducting research and/or implementing innovations using theoretical provisions and methods of system analysis K01. Ability to abstract thinking, analysis and synthesis. K03. Ability to conduct research at an appropriate level. K04. Ability to learn and master modern knowledge. K05. Ability to search, process and analyze information from various sources. K06. Ability to generate new ideas (creativity). K07. Ability to identify, pose and solve problems. professional competences: K13. Ability to develop and analyze mathematical models of natural, man-made, economic and social objects and processes. K14. Ability to plan and conduct system studies, perform mathematical and informational modeling of dynamic processes. K16. The ability to form new hypotheses and research tasks in the field of system analysis and decision-making, to choose appropriate directions for their application. K17. The ability to formulate, analyze and synthesize when solving scientific problems at an abstract level. K20. The ability to develop functions for forecasting the dynamics of the development of processes of a different nature in a deterministic and stochastic environment and to evaluate the quality of the forecast. K24. The ability to reveal situational and systemic uncertainties, to develop algorithms for overcoming conflicts. K26. The ability to self-educate and improve professional qualifications. Надіслати відгук Бічні панелі Історія Збережено
Learning outcomes: As a result of studying the academic discipline, the student must be able to demonstrate the following learning outcomes: PR01. Know and be able to apply in practice methods of system analysis, methods of mathematical and information modeling for building and researching models of objects and informatization processes. PR02. Know the methods of revealing uncertainties in the tasks of system analysis, be able to reveal situational uncertainties and uncertainties in the tasks of interaction, opposition and conflict of strategies, find a compromise when revealing conceptual uncertainty, etc. PR06. Know and be able to apply methods of evolutionary modeling and genetic methods of optimization, methods of inductive modeling and the mathematical apparatus of fuzzy logic, neural networks, game theory and distributed artificial intelligence, etc. PR08. Know and be able to identify (evaluate) the parameters of mathematical models of control objects in real time in the conditions of changes in its dynamics and the action of random disturbances, using measured signals of input and output coordinates of the object. PR10. Know the models, methods and algorithms of decision-making in conditions of conflict, unclear information, uncertainty and risk. PR11. Ability to search for information in specialized literature in the field of system analysis, using various resources: journals, databases, online resources. KOM1. Ability to communicate, including oral and written communication in Ukrainian and foreign languages ??(English, German, Italian, French, Spanish). KOM2. The ability to use a variety of methods, including modern information technologies, for effective communication at the professional and social levels. AiB1. Ability to adapt to new situations and make appropriate decisions. AiB2. The ability to realize the need for lifelong learning in order to deepen the acquired and acquire new professional knowledge. AiB3. The ability to take responsibility for the work performed, to make decisions independently, to achieve the set goal in compliance with the requirements of professional ethics. Надіслати відгук Бічні панелі Історія Збережено
Required prior and related subjects: Distributed information systems, Technologies for supporting decision-making processes, Recommendation systems.
Summary of the subject: The following topics are considered in the teaching of the discipline: Architecture of artificial neural networks. Neural network learning algorithms. Neural networks with feedback and self-organization. Fuzzy inference systems. Fuzzy neural networks and their use in prediction problems. The use of systems with fuzzy logic and fuzzy neural networks in forecasting problems in macroeconomics and financial analysis. Fuzzy neural networks in classification problems. Recognition of objects of electro-optical images using fuzzy neural networks. Method of inductive modeling in data mining problems. Cluster analysis in intelligent systems. Fuzzy cluster analysis algorithms. Genetic algorithms and evolutionary modeling. Evolutionary programming. Swarm algorithms of computational intelligence. Hybrid swarm optimization algorithms.
Опис: Architecture of artificial neural networks. Single-layer and multi-layer forward propagation neural networks. Recurrent networks without hidden and with hidden networks. Single-layer and multilayer perceptrons. Computing capabilities of neural networks. Neural networks with radial basis functions. Neural network learning algorithms. The gradient method of backpropagation neural network training. A gradient learning algorithm for a neural network with any number of layers and its improvement. A genetic algorithm for learning a neural network. Improving the convergence of neural network learning algorithms. Algorithm of conjugate gradients. Neural networks with feedback and self-organization. Hopfield neural network and its use. Hamming neural network. Architecture and work algorithm. Self-organizing neural networks. Kohonen's learning algorithms. Using Kohonen neural networks. Systems of fuzzy logical inference. Algorithms of fuzzy logic inference (Method of Mamdani and Tsukamoto). Methods of bringing to clarity. Universal approximation theorem. Fuzzy neural networks and their use in forecasting tasks. Mamdani and Tsukamoto inference fuzzy neural networks. Mamdani and Tsukamoto's gradient algorithm for learning fuzzy neural networks. ANFIS fuzzy network. Structure and learning algorithm. Neo-fuzzy neural networks and cascaded neo-fuzzy neural networks. The use of fuzzy logic systems and fuzzy neural networks in forecasting tasks in macroeconomics and financial analysis. Forecasting in macroeconomics and finance using fuzzy neural networks. Optimizing the investment portfolio in conditions of uncertainty. Forecasting the risk of bank bankruptcy under conditions of uncertainty based on fuzzy neural networks. Fuzzy neural networks in classification tasks. NEFClass fuzzy neural networks. Architecture, properties, learning algorithm, bases of rules and parameters of membership function of fuzzy sets. Analysis of NEFClass properties. Modified NEFClass-M fuzzy classification system. Comparative analysis of NEFClass and NEFClass-M fuzzy neural networks in economic classification problems. Object recognition of electro-optical images using fuzzy neural networks. General characteristics of the system. Types of sensors and hyperspectral systems. The use of the NEFClass system in the task of object recognition of electro-optical images with real data. The method of inductive modeling in tasks of intellectual data analysis. General characteristics and basic principles of the method of group consideration of arguments. Fuzzy method of group consideration of arguments (NMGUA). The main ideas of the method. Mathematical model of NMGUA. Description of the NMGUA algorithm. NMGUA with Gaussian and bell-shaped membership functions. Cluster analysis in intelligent systems. Cluster analysis. Setting the problem. Quality criteria and metrics of cluster analysis. Classification of cluster analysis algorithms. Fuzzy k-means method. Determining the initial location of cluster centers. Methods of peak and difference grouping. Fuzzy cluster analysis algorithms. Gustavson-Kessel fuzzy cluster analysis. Use of fuzzy k-means and Gustavson-Kessel methods in automatic classification problems. Genetic algorithms and evolutionary modeling Canonical genetic algorithm. Binary representation. Floating point representation. The use of the genetic algorithm in the problem of structural synthesis of networks. Evolutionary programming. Evolutionary programming operators. Finding the shortest path along a graph based on a genetic algorithm. Adaptation (self-learning) of evolutionary programming. Implementation of evolutionary programming algorithms. Надіслати відгук
Assessment methods and criteria: Diagnostics of knowledge takes place by evaluating the completed laboratory work and examination control (written and oral components) in the form of test questions.
Критерії оцінювання результатів навчання: • Current control (40%): written reports on laboratory work, oral examination; • Final control (60%). in written – 50%, verbally- 10%.
Порядок та критерії виставляння балів та оцінок: 100–88 points – (“excellent”) is awarded for a high level of knowledge (some inaccuracies are allowed) of the educational material of the component contained in the main and additional recommended literary sources, the ability to analyze the phenomena being studied, in their relationship and development, clearly, succinctly, logically, consistently answer the questions, the ability to apply theoretical provisions when solving practical problems; 87–71 points – (“good”) is awarded for a generally correct understanding of the educational material of the component, including calculations, reasoned answers to the questions posed, which, however, contain certain (insignificant) shortcomings, for the ability to apply theoretical provisions when solving practical tasks; 70 – 50 points – (“satisfactory”) is awarded for weak knowledge of the component’s educational material, inaccurate or poorly reasoned answers, with a violation of the sequence of presentation, for weak application of theoretical provisions when solving practical problems; 49–26 points – (“not certified” with the possibility of retaking the semester control) is awarded for ignorance of a significant part of the educational material of the component, significant errors in answering questions, inability to apply theoretical provisions when solving practical problems; 25-00 points - ("unsatisfactory" with mandatory re-study) is awarded for ignorance of a significant part of the educational material of the component, significant errors in answering questions, inability to navigate when solving practical problems, ignorance of the main fundamental provisions.
Recommended books: 1. Згуровський М. З., Зайченко Ю. П. Основи обчислювального інтелекту. – Київ: Науково-виробниче підприємство «Видавництво «Наукова думка» НАН України» 2013, 407 с. 2. Harrison Kinsley, Daniel Kukiela. Neural Networks from Scratch in Python. Sentdex, Kinsley Enterprises Inc. 2020, 658 p. https://nnfs.io/ 3. Зайченко Ю. П. Основи проектування інтелектуальних систем. Навчальний посібник. – К.: Видавничий Дім «Слово», 2004. – 352 с. 4. Руденко О. Г., Бодянський Є. В. Штучні нейронні мережі: Навчальний посібник. – Харків: ТОВ «Компанія СМІТ», 2006. – 404 с. 5. М.А. Новотарський, Б.Б. Нестеренко. Штучні нейронні мережі: обчислення // Праці Інституту математики НАН України. – Т50. – Київ: Ін-т математики НАН України, 2004. – 408 с. 6. Пелещак Р.М., Литвин В.В., Пелещак І.Р. Методичний посібник для лабораторних робіт з навчальної дисципліни «Обчислювальний інтелект», 2021, 70 с.
Уніфікований додаток: Lviv Polytechnic National University ensures the realization of the right of persons with disabilities to obtain higher education. Inclusive educational services are provided by the Service of accessibility to learning opportunities "Without restrictions", the purpose of which is to provide permanent individual support for the educational process of students with disabilities and chronic diseases. An important tool for the implementation of the inclusive educational policy at the University is the Program for improving the qualifications of scientific and pedagogical workers and teaching and support staff in the field of social inclusion and inclusive education. Contact at: St. Karpinsky, 2/4, 1st floor, room 112 E-mail: nolimits@lpnu.ua Websites: https://lpnu.ua/nolimits https://lpnu.ua/integration
Академічна доброчесність: The policy regarding the academic integrity of the participants in the educational process is formed on the basis of compliance with the principles of academic integrity, taking into account the norms "Regulations on academic integrity at the Lviv Polytechnic National University" (approved by the academic council of the university on June 20, 2017, protocol No. 35).