Scribd
Upload
129 views2 pages

AIML Assignment 2

This document outlines an assignment for the course 'Artificial Intelligence and Machine Learning' at PSNA College of Engineering & Technology for the academic year 2024-2025. It includes course objectives, a series of questions related to machine learning concepts, and specifies the assignment's structure and grading. The assignment covers topics such as ensemble learning, K-Nearest Neighbors, deep learning model training, and neural network activation functions.

Uploaded by

Gayathri Ramasamy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
129 views2 pages

AIML Assignment 2

This document outlines an assignment for the course 'Artificial Intelligence and Machine Learning' at PSNA College of Engineering & Technology for the academic year 2024-2025. It includes course objectives, a series of questions related to machine learning concepts, and specifies the assignment's structure and grading. The assignment covers topics such as ensemble learning, K-Nearest Neighbors, deep learning model training, and neural network activation functions.

Uploaded by

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

PSNA COLLEGE OF ENGINEERING & TECHNOLOGY, DINDIGUL -624622.

(An Autonomous Institution, Affiliated to Anna University Chennai)

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

Academic Year 2024 – 2025 (EVEN Semester)


Assignment – II

Course Code :EC2612 Course Name : Artificial Intelligence and Machine Learning
Degree : B.E Programme : ECE Semester & Sec : VI & A
Type of Assignment : Common Max. Marks : 40
Faculty In-charge : [Link] Date : 10.04.2025
Course Objectives
CO1 Use appropriate search algorithms for problem solving
CO2 Apply reasoning under uncertainty
CO3 Build supervised learning models
CO4 Build ensembling and unsupervised models
CO5 Build deep learning neural network models
Answer All questions. ( 4*10=40 Marks)
[Link]. Questions Mar C BL
k O
1 10 3 3

2 You are building a machine learning model for a financial institution to 10 4 3


detect credit card fraud, where fraudulent transactions make up less than
1% of the data. You’ve tested individual models: a Decision Tree catches
most fraud cases but triggers many false positives, while Logistic
Regression performs better on non-fraud cases but misses subtle fraud
patterns. Given the critical nature of detecting fraud with both high
precision and high recall, how can you apply ensemble learning
techniques to improve model performance in this highly imbalanced
scenario? Which ensemble method would you choose,
3 The table below represents data set. You have two columns 5 4 3
— Brightness and Saturation. Each row in the table has a class of
either Red or Blue. Assume the value of K is 5. Calculate Euclidean
Distance in the K-Nearest Neighbors Algorithm for the new data entry
(20,35).
Brightness Saturation Class
40 20 Red
50 50 Blue
60 90 Blue
10 25 Red
70 70 Blue
60 10 Red
25 80 Blue

4 You are tasked with training a deep learning model to predict housing 10 5 3
prices based on various features such as location, square footage,
number of bedrooms, etc. After training the model using Stochastic
Gradient Descent (SGD), you notice that the model's loss function is
fluctuating significantly during training, and the model performance
on the validation set is inconsistent. The learning rate is set to a
constant value. What could be causing these fluctuations in training,
and how would you modify the training process (e.g., learning rate,
optimization techniques) to improve the stability and performance of
the model?
5 You are building a neural network to classify images of cats and dogs. 5 5 3
You decide to use ReLU as the activation function for the hidden layers.
After training the model, you notice that it works well initially, but some
of the neurons in the network seem to stop responding (they output zero
for all inputs).What could be causing the neurons to output zero for all
inputs in your neural network using ReLU?

Faculty In-charge Course Coordinator HOD - CSE

You might also like