WITH EFFECT FROM THE ACADEMIC YEAR 2013 - 2014
ME 406
NEURAL NETWORKS
(ELECTIVE - I)
Instruction 4 Periods per week
Duration of University Examination 3 Hours
University Examination 75 Marks
Sessional 25 Marks
Unit-I
Introduction: Knowledge - based information processing. A general view of knowledge based algorithm. Neural information processing. Hybrid intelligence. Artificial neuron.
Unit-II
Basic Neural Computation Models: Basic concepts of Neural network - Network properties, node properties, sigmoid functions. System dynamics. Inference and learning algorithm. Data representation. Functional classification models - single layer perceptions. Multilayer perceptions.
Unit-III
Learning supervised and unsupervised statistical learning. AI learning. Neural network Learning - Back propagation algorithm and derivation. Stopping criteria. Complexity of Learning Generalization.
Unit-IV
Self- organizing Networks: Introduction, The Kohonen algorithm, weight initialization, weight training, associative memories, bi-directional associative memories.
Unit-V
Hopfield Networks: Introduction: The Hopfield model. Hopfield network algorithm. Boltzman’s machine algorithm. Neural applications.
Suggested Reading:
- 1.Limin Fu, Neural Networks in Computer Intelligence , Mc-Graw Hill, 1995.’
- 2.Bart Kosho, Neural Networks and Fuzzy Systems, Prentice Hall of lndia, 1994.
- 3.James A. Freeman, Simulating, Neural Networks, Addison Wesley Publications, 1995.