Scribd
Upload
14 views

qb os 2

The document outlines the structure and requirements for an in-semester examination on Deep Learning for AI&ML students, scheduled for April 26, 2024. It includes instructions for answering questions, which cover various topics such as ethical considerations, generative models, challenges in medical diagnosis, and Transformer-based models. Students are expected to demonstrate their understanding through a combination of theoretical questions and practical applications.

Uploaded by

dubeyvarun949
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
14 views

qb os 2

The document outlines the structure and requirements for an in-semester examination on Deep Learning for AI&ML students, scheduled for April 26, 2024. It includes instructions for answering questions, which cover various topics such as ethical considerations, generative models, challenges in medical diagnosis, and Transformer-based models. Students are expected to demonstrate their understanding through a combination of theoretical questions and practical applications.

Uploaded by

dubeyvarun949
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

IN-SEMESTER EXAMINATION-II

BT (Semester-VIII)
SUBJECT – Deep Learning
Branch: AI&ML Date: 26/04/2024
Div.: ALL Timing: 12:00 PM to 1:00 PM
Duration: 60 Minutes Maximum Marks: 20
Instructions –
1. All questions are compulsory.
2. Assume suitable data wherever necessary and state the assumptions made.
3. Diagrams / sketches should be given wherever necessary.
4. Use of logarithmic table, drawing instruments and non-programmable calculators is permitted.
5. Figures to the right indicate full marks.

Course
Q. Mark Learnin
1
Answer any 5 of the following questions s
Outcome
g Levels
s
Differentiate ethical considerations associated with deploying deep learning
models in nanotechnology, Discuss issues such as data privacy, bias in
a.
algorithmic decision-making, and potential societal impacts of
2 6 AN
nanotechnology-enabled applications
b Define generative models and briefly explain their purpose in machine
.
2 4 R
learning
List some challenges and limitations of deploying deep learning
c. 2 6 R
models for medical diagnosis in real-world clinical settings.
d Identify how attention mechanisms influence the architecture of
.
2 5 R
Transformer-based Large Language Models
State recurrent neural network(RNN) can model complex temporal
e. 2 6 A
dynamics in nanoscale systems.

f. List challenge associated with training Variational Autoencoders (VAEs) 2 4 U


g List the role of positional encoding in Transformer-based LLMs and its
.
2 5 R
importance in capturing the sequential nature of language
Develop a system capable of translating images from one domain to
another, such as converting sketches into realistic photographs. Propose a
Q.
a. novel approach that combines elements of Variational Autoencoders 5 4 A
2
(VAEs) and Generative Adversarial Networks (GANs) to achieve this task,
outlining the potential benefits and challenges of your proposed method.
OR
Brief the Generative Adversarial Networks. how do they work? Explain the
b
concept of adversarial training between the generator and discriminator 5 4 A
.
networks.
Q. Discuss the importance of vocabulary size and tokenization techniques in
a. 5 5 A
3 designing the input layer of Transformer-based LLMs
OR
Design a Transformer-based Large Language Model (LLM), consider
various components to influence the architecture. Give sentiment analysis
b
.
on social media posts, discuss how you would prioritize and integrate 5 5 AN
specific components to enhance the model's ability to understand and
classify sentiment accurately.

You might also like