MD3315 NEURAL NETWORKS AND APPLICATIONS SYLLABUS | ANNA UNIVERSITY BE MEDICAL ELECTRONICS ENGINEERING 6TH SEMESTER SYLLABUS REGULATION 2008 2011 2012-2013 BELOW IS THE ANNA UNIVERSITY SIXTH SEMESTER B.E MEDICAL ELECTRONICS ENGINEERING DEPARTMENT SYLLABUS, TEXTBOOKS, REFERENCE BOOKS,EXAM PORTIONS,QUESTION BANK,PREVIOUS YEAR QUESTION PAPERS,MODEL QUESTION PAPERS, CLASS NOTES, IMPORTANT 2 MARKS, 8 MARKS, 16 MARKS TOPICS. IT IS APPLICABLE FOR ALL STUDENTS ADMITTED IN THE YEAR 2011 2012-2013 (ANNA UNIVERSITY CHENNAI,TRICHY,MADURAI, TIRUNELVELI,COIMBATORE), 2008 REGULATION OF ANNA UNIVERSITY CHENNAI AND STUDENTS ADMITTED IN ANNA UNIVERSITY CHENNAI DURING 2009
MD3315 NEURAL NETWORKS AND APPLICATIONS L T P C
3 1 0 4
UNIT I NEURON MODEL NETWORK ARCHITECTURE 9
Neuron model – single input neuron –activation function – multiple input neuron neural
networks viewed as directed graphs -feedback - network architectures – knowledge
representation – linear and non- linear separable problem(XOR)
UNIT II LEARNING PROCESS 9
Error – correction learning – memory based learning - hebbian learning-competitive
learning-Boltzmann learning-credit assignment problem-supervised and unsupervised
learning-adaptation-statistical learning theory.
50
UNIT III PERCEPTIONS 9
Single layer perception-Adaptive filtering-unconstrained optimization-Least-mean square
algorithm-Leaning curve-Annealing Technique-perception convergence theorem-
Relationship between perception and Baye’s classifier-Back propagation algorithm-
Network pruning techniques-supervised learning viewed as an optimization problemconvolutional
network. Application to Adaptive Prediction and character recognition.
UNIT IV ATTRACTOR NEURAL NETWORK AND ART 9
Hopfield model-BAM model-BAM stability-Adaptive BAM -Lyapunov function-effect of
gain-Hopfield design-Application to TSP problem-ART- layer 1-layer 2-orienting
subsystem-Leaning lawL1-L2-Leaning law L2-L1-ART algorithm-ARTMAP
UNIT V PRINCIPAL COMPONENT ANALYSIS AND SELF ORGANIZATION 9
Principle of self organization-Principle Component analysis-Adaptive PCA using Lateral
inhibition-Two classes of PCA algorithm-Two basic feature- mapping model-self
organizing map-SOM Algorithm-properties of the feature map-LVQ-Hierarchical vector
Quantization. Applications of self-organizing maps: The Neural Phonetic Typewriter-
Learning Ballistic Arm Movements
TUTORIAL : 15 TOTAL: 60 PERIODS
REFERENCES:
1. Simon Haykin, “Neural Networks and Learning Machines” -3/E - Pearson/ Prentice
Hall 2009
2. Satish Kumar- “Neural Networks : A Classroom Approach”-TMH-2008
3. Freeman J.A., Skapura D.M.”Neural networks, algorithms, applications, and
programming techniques” -Addition Wesley, 2005.
4. Laurene Fausett, “Fundamentals of Neural Networks: Architectures, Algorithms,
and Applications” - Pearson/ Prentice Hall
5. Robert J Schalkoff-“Artificial Neural Networks,McGraw Hill”-1997
* Common with ECE
MD3315 NEURAL NETWORKS AND APPLICATIONS L T P C
3 1 0 4
UNIT I NEURON MODEL NETWORK ARCHITECTURE 9
Neuron model – single input neuron –activation function – multiple input neuron neural
networks viewed as directed graphs -feedback - network architectures – knowledge
representation – linear and non- linear separable problem(XOR)
UNIT II LEARNING PROCESS 9
Error – correction learning – memory based learning - hebbian learning-competitive
learning-Boltzmann learning-credit assignment problem-supervised and unsupervised
learning-adaptation-statistical learning theory.
50
UNIT III PERCEPTIONS 9
Single layer perception-Adaptive filtering-unconstrained optimization-Least-mean square
algorithm-Leaning curve-Annealing Technique-perception convergence theorem-
Relationship between perception and Baye’s classifier-Back propagation algorithm-
Network pruning techniques-supervised learning viewed as an optimization problemconvolutional
network. Application to Adaptive Prediction and character recognition.
UNIT IV ATTRACTOR NEURAL NETWORK AND ART 9
Hopfield model-BAM model-BAM stability-Adaptive BAM -Lyapunov function-effect of
gain-Hopfield design-Application to TSP problem-ART- layer 1-layer 2-orienting
subsystem-Leaning lawL1-L2-Leaning law L2-L1-ART algorithm-ARTMAP
UNIT V PRINCIPAL COMPONENT ANALYSIS AND SELF ORGANIZATION 9
Principle of self organization-Principle Component analysis-Adaptive PCA using Lateral
inhibition-Two classes of PCA algorithm-Two basic feature- mapping model-self
organizing map-SOM Algorithm-properties of the feature map-LVQ-Hierarchical vector
Quantization. Applications of self-organizing maps: The Neural Phonetic Typewriter-
Learning Ballistic Arm Movements
TUTORIAL : 15 TOTAL: 60 PERIODS
REFERENCES:
1. Simon Haykin, “Neural Networks and Learning Machines” -3/E - Pearson/ Prentice
Hall 2009
2. Satish Kumar- “Neural Networks : A Classroom Approach”-TMH-2008
3. Freeman J.A., Skapura D.M.”Neural networks, algorithms, applications, and
programming techniques” -Addition Wesley, 2005.
4. Laurene Fausett, “Fundamentals of Neural Networks: Architectures, Algorithms,
and Applications” - Pearson/ Prentice Hall
5. Robert J Schalkoff-“Artificial Neural Networks,McGraw Hill”-1997
* Common with ECE
No comments:
Post a Comment
Any doubt ??? Just throw it Here...