De Gruyter Fuzzy Machine Learning: Advanced Approaches to Solve Optimization Problems

The continuous increase in demand towards lowering production costs to withstand cutthroat competition has prompted researchers to look for rigorous methods of decision making. As search for best has always fascinated mankind, operations or strategies have been attempted and devised for searching optimum solutions for variety of problems in all branches of activities perceived by logic or intitution or both. With this motivation in this work novel fuzzy machine learning algorithms are proposed to develop optimum solutions for several optimization problems. The emphasis of proposed methodologies is given on handling data sets which are large both in size and dimension and involves classes that are overlapping, intractable and have non-linear boundaries. Several strategies based on data reduction, dimensionality reduction, active learning efficient search heuristics are employed for dealing with scaling issues The problems handle linguistic input and ambiguous output decision, learning of overlapping and intractable class structures, selection of optimal parameters and discovering human comprehensible knowledge in form of linguistic rules. The different features of methodologies along with comparisons with those of related ones are demonstrated extensively on different real life data sets. The experimental data have been considered from variety of domains. The superiority of models over the benchmark are found to be effective and significant. This book is generally suited for bachelor, master and research students working for courses in engineering optimization, operations research and operations management. It will also be immensely beneficial to industry professions as well as to scientists and postdocs working at research institutes working in abovementioned domains. The prima face of this book is to highlight certain insights and findings for some important optimization problems over the period of past 10 years. All the proposed algorithms are of continuous evolving nature. The problems considered here include traveling salesman problem, transportation problem, non-convex optimization, dynamic programming, strategic games, resource allocation problem and job assignment problems. The solutions of these problems are developed using machine learning approaches by integrating fuzzy logic, neural networks, deep learning, and genetic algorithms. Table of contents Introduction Mathematical Foundations: Optimization and Fuzzy Machine Learning Hieararchical Fuzzy Self Organizing Map for Traveling Salesman Problem Transportation Problem through Probalistic and Fuzzy Uncertainties Hierarchical Fuzzy Deep Learning for Strategic Game Theory Problem Fuzziness in Non-Convex Optimization Fuzziness in Dynamic Programming Fuzzy Multi-Objective Optimization for Resource Allocation Problem Hierarchical Fuzzy Deep Learning for Job Assignment Problems