Course Archives Statistical Quality Control & Operations Research Unit
Course:Neural Network
Level: Postgraduate
Time: Currently not offered
Syllabus
Past Exams


Syllabus: Introduction to Soft Computing and ANN: An overview of analysis and Design of intelligent systems using soft computing techniques, Basic Concepts of Artificial Neural Network (ANN), Similarity with biological neurons, General characteristics, Historical development and domain specific applications, Statistical modeling and ANN.
Building blocks of ANN and Fundamental ANN Models: Architecture, Weights, Bias, Net Input, Threshold, Activation functions, Training and its related parameters, Simulation. McCulloch-Pitts and Hebb Nets: architecture and algorithms with examples.
Learning Rules of ANN: Hebbian, Perceptron, Delta, Competitive, Perceptron convergence theorem.
Typical Networks: Single Layer Perceptron architecture, training and application algorithm. Adaline and Madaline - architecture, training and application algorithm Discrete Hopfield Net - architecture, training and application algorithm.
Feed Forward Networks: Multi Layer Perceptron (MLP) Generalized Delta (Back Propagation) Learning rule, architecture, training algorithm, selection of parameters, learning constraints, application algorithm, local optimum, merits and demerits, applications.
Radial Basis Function (RBF) architecture, training algorithm.
Self Organizing Map: Kohonen Self Organizing Feature Maps (SOM) - architecture, training algorithm. Learning Vector Quantization (LVQ) - architecture, training algorithm.
Some Special Purpose Networks: Ensemble networks purpose and concepts. Adaptive Resonance Theory (ART) - architecture, training algorithm, ART 1, ART 2. Probabilistic Neural Network (PNN) - architecture, training algorithm. Modular Networks. Case Studies.
Use of Software: Developing ANN models with the help of computer software such as MATLAB, STATISTICA, NEUROMAT etc. for solving real-life problems and related performance measures with graphical interface.


Reference Texts:
1. An Introduction to Neural Networks: By K. Gurney, UCL Press.
2. Computational Intelligence: By Andries P Engelbrecht, John Wiley & Sons Ltd., 2003.
3. Neural Networks, Fuzzy Logic and Genetic Algorithms: By S. Rajasekaran and G.A.V.Pai, PHI.
4. Neural Network Fundamentals, 1996: By N.K. Bose, P. Liung:, McGraw Hill Inc.,
5. Neural Networks, 1994: By Haykin Simon, Macmilan, U.K
6. Neural networks for pattern recognition, 1995: By Bishop, C., Oxford Univ. press
7. Neural networks for statistical modeling: By Murray Smith.
8. Neuro-Fuzzy PR-Methods in Soft Computing: By Sankar K.Pal and Sushmita Mitra, John Wiley & Sons


Top of the page

Past Exams
Midterm
Semestral
Supplementary and Back Paper

Top of the page

[Indian Statistical Institute]