Trending ▼   ResFinder  

2003 Course Artificial Neural Networks

4 pages, 45 questions, 0 questions with responses, 0 total responses,    0    0
pune_eng
  
+Fave Message
 Home > pune_eng >

Instantly get Model Answers to questions on this ResPaper. Try now!
NEW ResPaper Exclusive!

Formatting page ...

Total No. of Questions : 12] [Total No. of Pages : 4 [3764]-228 P1347 B.E. (E & T C) ARTIFICIAL NEURAL NETWORKS (404218) (2003 Course) Time : 3 Hours] [Max. Marks : 100 Instructions to the candidates : 1) Answer Q.1 or Q.2, Q.3 or Q.4, Q.5 or Q.6 from Section I. Q.7 or Q.8, Q.9 or Q.10, Q11 or Q12 from Section II. 2) Answers to the two sections should be written in separate answer books. 3) Neat diagrams must be drawn wherever necessary. 4) Use of electronic pocket calculator is allowed. 5) Assume suitable data, if necessary. SECTION - I Q1) a) Compare the performance of a computer and that of a biological neural network in terms of speed of processing, size and complexity, storage, fault tolerance and control mechanism. [6] b) Compare LMS, perceptron and delta learning laws. c) A 2 input neurons with the following parameter is given : = 1 .2, W = [3, 2] and I = [ 5, 6]. Calculate the neuron output for the following transfer function. [6] i) Symmetrical hard limit function. ii) [6] Saturating linear function. OR Q2) a) What are the main differences among the three modes of artificial neuron, namely, McCulloch-Pitts, perceptron and Adaline? [6] b) What is reinforcement learning? In what way is it different from supervised learning? [6] c) What is meant by operating range of neuron? [3] d) Explain what is Hebbian learning. [3] Draw the architecture of multilayer feedforward network. [4] Q3) a) P.T.O. b) Explain the training of multilayer feedforward networks by BPN algorithm. [8] c) Show that a multilayer network with linear transfer functions is equivalent to a single layer linear network. [4] OR Q4) a) b) Explain the architecture of adaline. Explain the training algorithm used in adaline. [8] Given I1 = [1 1]T, t1 = 1 I2 = [1 1]T, t2 = 1 represents input and target pairs, train the network using the delta rule with initial weight set to zero and a learning rate of 0.5. [4] c) Q5) a) How is Madaline different from adaline? Explain with diagrams. [4] Describe the Boltzmann machine. [4] b) Distinguish between clamped and free running conditions in Boltzmann machine during learning. [6] c) How to perform the following tasks by a Boltzmann machine? i) Pattern completion. ii) [6] Pattern association. OR Q6) a) Explain the term feedback network. [4] b) Explain the architecture and training algorithm used in Hopfield network. [8] c) Explain the term energy function. Explain the energy function associated with Hopfield network. [4] SECTION - II Q7) a) b) [3764]-228 Draw the architecture of ARTI network. Explain the same. [6] Explain the training algorithm in ART with significance of vigilance parameter. [8] 2 c) For the Kohonen network with two cluster units and five input units, find the new weight vector for the winning unit. Given initial weight, W = 1.0 0 . 5 0.6 0 . 4 0.2 0 . 2 0 . 4 0 . 6 0 . 8 1.0 If input pattern given is X = [0.5 1.0 0.5 0.0 0.0] and learning rate of 0.2. [4] OR Q8) a) Draw the architecture of Maxnet and explain. [4] b) How does Maxnet act as subnet. [3] c) Compare SOFM with LVQ. [6] d) What is Hamming net? [3] e) What is plasticity with reference to neural network? [2] Q9) a) Explain the following with reference to memory in Artificial neural networks: [6] i) Transient memory. ii) Temporary memory. iii) Short time memory. iv) Long term memory. b) What is an associative memory? What are the requirements of an associative memory? [4] c) What is a recurrent autoassociative memory? d) How is noise suppression achieved in a recurrent autoassociative memory? [3] [3] OR Q10)a) [3764]-228 Explain BAM architecture and explain the difference between BAM and other neural network architectures. [6] 3 b) Explain the distinction between: [4] Pattern association and pattern classification tasks. c) Give the architecture of Radial basis function network. [6] What is the problem in the recognition of handwritten digits? [5] b) What is image segmentation problem? [5] c) What is time delay neural network architecture? How is it suitable for classification of CV segments? [6] Q11)a) OR Q12)a) Explain how neural network principles are useful in control applications. [6] b) Explain the difficulties in the solution of travelling salesman problem by a feedback neural network. [5] c) What is the significance of neural networks in the NET talk application.[5] Y [3764]-228 4

Formatting page ...

Formatting page ...

Formatting page ...

 

  Print intermediate debugging step

Show debugging info


 


Tags : Pune, Engineering, University of Pune, Engineering question papers, Pune University, previous year question papers, question papers, india, model question paper, pune university paper pattern, pune university syllabus, old question papers  

© 2010 - 2025 ResPaper. Terms of ServiceContact Us Advertise with us

 

pune_eng chat