Course Interaction: Every Thursday (L10) 12:00-13:15

Course Contents and Lecture Wise Schedule

Topic

Details

Lecture Hours

Feed-forward Networks

Simple Neuron Model, Multi-layered Networks, Back-propagation Algorithm, Generalized delta rule, Radial basis function network, Adaptive Learning rate, Examples

8

Feedback Networks

Back-propagation through time, real-time recurrent learning, LSTM

6

Self-organizing Networks

Unsupervised Learning, Kohonen SOM, Extended Kohonen SOM

3

Application I

Visual Motor Coordination - learning to manipulate

3

Application II

Indirect and Direct Adaptive Control - learning that guarantees stability

5

Approximate Dynamic Programming

Adaptive Critic networks and learning approaches

6

Deep Learning

RBM, CNN, Deep Reinforcement Learning, Auto-encoder based deep network

12

 

Evaluation Components

Continuous Evaluation: 30%

Assignment Tests: 30%

Final Exam: 40%

Text Book

Laxmidhar Behera and I Kar, Intelligent Systems and Control, OUP, 2009, 5th Reprint

Ian Goodfellow, Yoshua Bengio and Aaron Courville, Deep Learning, www.deeplearningbook.org

Reference Book

Simon Haykin, Neural Networks - A Comprehensive Foundations, 1994