Scribd
Upload
22 views1 page

QB Dl-Cie1

Uploaded by

Harsh Kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
22 views1 page

QB Dl-Cie1

Uploaded by

Harsh Kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

UNIT_1

1. What is Machine intelligence? How machine learning differs from rule based
programming?
2. Define Deep learning. Mention the specific features and applications of Deep learning.
3. Differentiate between supervised and unsupervised machine learning.
4. Analyze the measures of central tendency and measures of dispersion in statistical
analysis of data distribution.
5. Briefly explain simple linear regression and illustrate the objective function.
6. How to identify the non linear trend in the data and decide upon the polynomial
regression? With an example code show how the feature vectors are transformed to
polynomial features.
7. Analyze the training flow for regression and classification (kNN) machine learning
techniques.
8. Discuss the utility of the following libraries: Numpy, Scipy, Pandas, Matplotlib, Seaborn,
Tensorflow, Keras etc.
9. Discuss the working of K-Means clustering algorithm and list out various distance
measures used in clustering algorithms.
10. Problems on linear regression, kMeans clustering and kNN algorithms..

UNIT_2
11. Given two neurons N1 and N2 representing hidden layer 1 and 2 are connected in a feed
forward network. The input ‘x’=0.1 is applied to N1 and actual output of N2 is 0.25.
Weights and biases at N1 and N2 are initialized as: w1=0.15, b1=0.4; w2=0.45, b2=0.65
and learning rate =0.4. Apply sigmoid activation and calculate mean squared error as the
loss function. Using GD, update w1 and w2 and analyze the condition of vanishing
gradient.
12. What is the use of activation function in a neural network? Analyze with illustration
various activation functions mentioning their features and utilities.
13. What is a perceptron? Dicuss the analogy between biological neuron and perceptron.
14. Illustrate the multilayer perceptron model and briefly discuss on the structure.
15. Design Boolean AND, OR, NOT, NAND, NOR logic using a single Perceptron and Ex-
OR/NOR logic using MLP with appropriate justification.
16. Explain SGD algorithm for multi layer networks illustrating how weights are updated
with back propagation.
17. Discuss on the loss functions used for classification and regression models.
18. Discuss on various evaluation metrics used for classification and regression models.
19. Given in the table, the scores for three classes at the output layer of a
classifier. Apply sigmoid/softmax activations individually. If the one-
hot encoded output is [0,1,0], calculate the cross entropy loss.

class Scores
1 1.2
2 2.0
3 -0.4
20. Analyze the techniques: Batch normalization and Dropout to manage the issue of overfitting of the model.
21. Initializers: Xavier Glorot and He: Normal and Uniform.
22. Programs on model training..

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy