Scribd
Upload
16 views2 pages

Deep Learning Assignment: Optimization & RNNs

The document outlines an assignment for the Deep Learning course at Vijaya Vittala Institute of Technology, covering various topics in artificial intelligence and data science. It includes questions on learning versus optimization, challenges in neural network optimization, parameter initialization strategies, and convolutional neural networks, as well as recurrent neural networks (RNNs) and performance metrics. The assignment is structured into modules, each focusing on different aspects of deep learning and neural networks.

Uploaded by

Kalpana Murthy
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
16 views2 pages

Deep Learning Assignment: Optimization & RNNs

The document outlines an assignment for the Deep Learning course at Vijaya Vittala Institute of Technology, covering various topics in artificial intelligence and data science. It includes questions on learning versus optimization, challenges in neural network optimization, parameter initialization strategies, and convolutional neural networks, as well as recurrent neural networks (RNNs) and performance metrics. The assignment is structured into modules, each focusing on different aspects of deep learning and neural networks.

Uploaded by

Kalpana Murthy
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

VIJAYA VITTALA INSTITUTE OF TECHNOLOGY

# 35/1, Dodda Gubbi Post,Hennur- Bagalur Road, Bengaluru - 560077.


Approved by AICTE, Affiliated to VTU & Accredited by NAAC
Department Of Artificial Intelligence and Data science
Deep learning (MCS105A)

Assignment 2
Module 3

1. Explain How Learning Differs from Pure Optimization.

2. Discuss the Challenges in Neural Network Optimization.

3. What are Parameter Initialization Strategies explain.

4. Explain stochastic gradient descent (SGT) algorithm.

5. How to choose right optimization algorithm.

6. Explain Optimization strategies & meta –algorithms.

7. Explain Convolution operation without kernel flipping in 2D.

8. Explain the components of convolutional neural network layer with neat diagram.

9. Discuss and mention examples of different format of data that can be used with convolutional
networks.

Module 4

10. Define RNNs, draw the computational graph to compute training loss of a RNN.

11. Illustrate the concept of teacher forcing which is applicable to RNNs.

12. Explain the common ways to provide inputs to an RNN with examples.

13. Explain Bidirectional RNNs.

14. Explain sequence to sequence RNN architecture.

15. Explain three important computations of most RNNs.

16. What are the challenges of long term dependencies explain briefly.
Module 5

17. Explain how to choose performance metrics and goals.

18. Discuss default baseline models.

19. Mention and explain 2 basic approaches for hyperparameters.

20. Explain alternative search method to grid search .

21. Explain Multi digit number recognition.

22. Mention and explain applications of NLP.

You might also like