Soft Computing (IT-701) - Important Questions

Describe the McCulloch-Pitts neuron model, its working, learning ability, and how it implements simple logic functions.

Define linear separability and explain why the XOR problem cannot be solved using a single-layer perceptron.

Compare ADALINE and MADALINE networks with respect to architecture and learning.

Explain the architecture of a single-layer feed-forward neural network / perceptron model with diagram and activation function.

Explain the backpropagation algorithm and its phases for training multilayer neural networks.

Explain the perceptron learning rule and training algorithm for a single-output perceptron, including its convergence characteristics.

Explain the Kohonen Self-Organizing Map (SOM) network, its architecture and working processes/algorithm with neat diagram and flowchart.

Explain the architecture and algorithm of the (forward-only) counterpropagation network with diagram.

Differentiate between Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) with application examples.

Explain the properties and set-theoretic operations of fuzzy sets with examples.

Construct a Fuzzy Inference System (FIS) for temperature control using appropriate fuzzy rules.

List and explain applications of fuzzy logic in engineering problems.

Explain the working of Genetic Algorithm with flowchart and discuss terminology and operators - individual, gene, fitness, population, encoding, selection, crossover, mutation and convergence criteria.

Describe the Schema Theorem and explain how it supports the working of genetic algorithms.

Write short notes on any two: (i) Encoding schemes used in genetic algorithm (ii) Ant Colony Optimization (iii) Fuzzy if-then rules and fuzzy reasoning.