|
|
|
General Course Information: Course Credits : 04 Assessment : Periodical Tests, Assignments, Semester Examination Contact Hours : 04 Periods per week Meeting Times : 1.45 P.M. to 3.15 P.M., Tuesday & Wednesday. Location : Computer Centre Classroom
Course Objectives & Prerequisites: Soft Computing refers to a collection of computational techniques in computer science, artificial intelligence and engineering disciplines which attempt to study, model and analyze complex problems - those for which more conventional methods have not yielded low cost, analytic and complete solutions. Unlike conventional computing, soft computing techniques are tolerant of imprecision, uncertainty and approximations. A typical course in soft computing generally comprises of computational techniques like Genetic/ Evolutionary algorithms, Artificial Neural Networks, Fuzzy Systems, Machine learning and probabilistic reasoning etc. This course thoroughly discusses Genetic Algorithms, Artificial Neural Networks (major topologies and learning algorithms) and Fuzzy Logic. Throughout the course the focus is on computational and applied aspects. By the end of the course a student is expected to become able to apply Genetic Algorithms and Artificial Neural Networks as computational tools to solve a variety of problems in their area of interest ranging from Optimization problems to Pattern recognition and control tasks. The prerequisite for this course is a basic understanding of problem solving, design and analysis of algorithms and computer programming. A prior course in Artificial Intelligence will be an advantage.
Course Syllabus: Introduction to Genetic Algorithm, Genetic Operators and Parameters, Genetic Algorithms in Problem Solving, Theoretical Foundations of Genetic Algorithms, Implementation Issues.
Neural Model and Network Architectures, Perceptron Learning, Supervised Hebbian Learning, Backpropagation, Associative Learning, Competitive Networks, Hopfield Network, Computing with Neural Nets and applications of Neural Network.
Introduction to Fuzzy Sets, Operations on Fuzzy sets, Fuzzy Relations, Fuzzy Measures, Applications of Fuzzy Set Theory to different branches of Science and Engineering.
Text & Reference Books:
(1) M. Mitchell, An Introduction to Genetic Algorithms, Prentice-Hall, 1998. (2) D. E. Goldberg, Genetic Algorithms in Search, Optimization, and Machine Learning, Addison-Wesley, 1989. (3) S. V. Kartalopoulos, Understanding Neural Networks and Fuzzy Logic: Basic Concepts and Applications, IEEE Press - PHI, 2004. (4) S. Rajasekaran & G. A. Vijayalakshmi Pai, Neural Networks, Fuzzy Logic and Genetic Algorithms: Synthesis & Applications, PHI, 2003. (5) S. N. Sivanandam & S. N. Deepa, Principles of Soft Computing, Wiley - India, 2007.
Web Resources: A number of resources on the web can be referred to for quality material on Soft Computing Techniques. The area is relatively new & rapidly changing one therefore, Students are advised to refer to regularly these resources for latest work/ developments. Few important journals and portals reporting work in the area of Soft Computing are: · IEEE Computational Intelligence Society · IEEE Transactions on Evolutionary Computation · IEEE Transactions on Neural Networks · ACM Special Interest Group on Genetic and Evolutionary Computation · Genetic and Evolutionary Computation Conferences · Springer Neural Computing and Applications · Elsevier Applied Soft Computing · IEEE Systems, Man and Cybernetics Society · IEEE Transactions on Fuzzy Systems Course Topics & Suggested Readings: This section contains detailed course topics, Suggested readings for all topics, Lecture slides/ handouts, local copies of selected references and links to other websites of interest. Access to this section is restricted to authorized persons only. You need a valid user id and password (to be obtained from the instructor in person or through Email) to access this section. Authorized users Please Click here to access.
Assignments: Every student is required to do at least one problem assignment involving solving a problem of interest using either Genetic Algorithm or Artificial Neural Network formulation. MATLAB and JAVA are the two programming languages to be used to implement the algorithmic solution. Here is an indicative list of problems. Evaluation: In addition to continuous assessment through Periodical tests and assignments there will be a semester end examination (70% weightage) at the end of the course. Here are question papers for previous year’s semester end examination. |
|
Department of Computer Science, Banaras Hindu University, Varanasi. |
|
(Elective - Odd Semesters) ( Updated July 2008) |
