Neural Networks and Applications

Sengupta, S. (Somnath)


Modules / Lectures : Introduction to Artificial Neural Networks ; Artificial Neuron Model and Linear Regression ; Gradient Descent Algorithm ; Nonlinear Activation Units and Learning Mechanisms ; Learning Mechanisms-Hebbian,Competitive,Boltzmann ; Associative memory ; Associative Memory Model ; Condition for Perfect Recall in Associative Memory ; Statistical Aspects of Learning ; V.C. Dimensions: Typical Examples ; Importance of V.C. Dimensions Structural Risk Minimization ; Single-Layer Perceptions ; Unconstrained Optimization: Gauss-Newton's Method ; Linear Least Squares Filters ; Least Mean Squares Algorithm ; Perceptron Convergence Theorem ; Bayes Classifier & Perceptron: An Analogy ; Bayes Classifier for Gaussian Distribution ; Back Propagation Algorithm ; Practical Consideration in Back Propagation Algorithm ; Solution of Non-Linearly Separable Problems Using MLP ; Heuristics For Back-Propagation ; Multi-Class Classification Using Multi-layered Perceptrons ; Radial Basis Function Networks: Cover's Theorem ; Radial Basis Function Networks: Separability & Interpolation ; Posed Surface Reconstruction ; Solution of Regularization Equation: Greens Function ; Use of Greens Function in Regularization Networks ; Regularization Networks and Generalized RBF ; Comparison Between MLP and RBF ; Learning Mechanisms in RBF ; Introduction to Principal Components and Analysis ; Dimensionality reduction Using PCA ; Hebbian-Based Principal Component Analysis ; Introduction to Self Organizing Maps ; Cooperative and Adaptive Processes in SOM ; Vector-Quantization Using SOM ;

Published by:

National Programme on Technology Enhanced Learning (NPTEL)

DOER Persistent Identifier: