Final QB DLT Past Paper PDF

LAQS: 1. Illustrate how deep learning fits within representation learning using a Venn diagram. 12 M 2. Explain representation learning methods. 3. **Describe the anatomy of a neural network with a clear diagram. 12 M \*\*\*\*\*** 4. Explain the process of building intelligent machin...

LAQS: 1. Illustrate how deep learning fits within representation learning using a Venn diagram. 12 M 2. Explain representation learning methods. 3. **Describe the anatomy of a neural network with a clear diagram. 12 M \*\*\*\*\*** 4. Explain the process of building intelligent machines. 5. What role does the learning rate play in the delta rule? 6. What is the purpose of dividing a dataset into a train set and a test set? 7. Explain different major architectures of deep networks. 8. Discuss about Recursive Neural Network. 9. What are the shortcomings of using feature selection techniques in neural networks. 10. Discuss about vanilla deep neural networks. 11. What is the difference between artificial intelligence, machine learning, and deep learning? 12. What achievements has deep learning made so far? 13. Demonstrate the different activation functions used in modeling artificial neurons. 14. How do you configure the learning rate for training a neural network? 15. How does gradient descent optimize the weights in a neural network that uses sigmoid neurons? 16. How does backpropagation compute gradients for a neural network with sigmoid activation? 17. Explain Unsupervised Pretrained Networks? 18. Discuss Recurrent Neural Networks in detail? 19. What is the purpose of using max pooling in CNNs. 20. What is the typical architecture of a convolutional neural network (CNN) used for classifying the MNIST dataset? 21. What is a linear neuron and write their limitations. 22. How does Mini-Batch Gradient Descent improve upon the limitations of Stochastic Gradient Descent? 23. **List and explain the historical trends in deep learning.** 24. **Analyse and write short notes on data augmentation?** 25. **What is the relationship between receptive fields of neurons in the human visual cortex and the receptive of CNN? 12 M** **SAQS:** 1. What is deep learning? 2. What are some common loss functions used in deep learning. 3. What is the process of training a feedforward neural network? 4. What is Recurrent Neural Network? 5. How do neural networks mimic the human visual system. 6. What is the role of layers in a neural network. 7. State the perceptron rule. 8. What is delta rule in gradient descent? 9. Describe the shape or dimension of input dad for recurrent neural network. 10. Distinguish between biological and artificial neurons? 11. What is the future promise of artificial intelligence? 12. What are the building blocks of deep learning. 13. What is gradient descent. 14. What are the major types of deep learning architectures? 15. What is Batch Normalization in CNN? 16. What is an optimizer in deep learning, and how does it work?