CourseDiary_CST395 - NEURAL NETWORKS AND DEEP LEARNING.pdf

vvcx

1. Subject CST395 - NEURAL NETWORKS AND DEEP LEARNING
Batch CS 2K20
Academic Year 2022-23
Total hours taken 19
Name of Teacher ASJAD NABEEL P
Designation Asst. Professor
Department Computer Science and Engineering
GOVERNMENT COLLEGE OF ENGINEERING KANNUR
MANGATTUPARAMBA PARASSINIKKADAVU.P.O KANNUR-670563
PH: 04972780227

COURSE DIARY
CST395 - NEURAL NETWORKS AND DEEP LEARNING
1

2. Sl.No. Content
1 GENERAL INSTRUCTIONS
2 VISION & MISION OF COLLEGE
3 PROGRAM OUTCOME
4 VISION & MISSION OF DEPARTMENT
5 PROGRAM EDUCATIONAL OBJECTIVES
6 PROGRAM SPECIFIC OUTCOMES
7 COURSE OUTCOMES
8 ACADEMIC CALENDAR
9 TIMETABLE
10 SYLLABUS
11 SUBJECT PLAN
12 SUBJECT COVERAGE
13 ASSIGNMENTS
14 SERIES EXAMS
15 INTERNALMARK
16 COURSE EXIT SURVEY
17 YEAR CALENDAR
18 ATTENDANCE SUMMARY
TABLE OF CONTENTS
CST395 - NEURAL NETWORKS AND DEEP LEARNING
2

3. General Instructions
Student's performance should be evaluated solely on an academic basis
Student's evaluation should be fair, consistent, transparent and accountable.
Evaluation of student's performance should be disclosed to the students.
1. Keep the Course Diary up to date by clearly indicating the subject coverage and students attendance on the relevant pages.
2. Paste the syllabus in the relevant page.
3. Write / paste the Course plan in the relevant page.
4. Events in a semester such as Series Test days, Cultural / Celebration days, days for extra / co-curricular activities etc. may be indicated in the Year Calendar.
5. Assignment details may be written in the Course Diary or may be filed in the Course File.
(i) Minimum 2 no. of assignments should be given.
(ii) Different sets of questions may be given in an assignment (atleast three) to a class.
(iii) Assignments may be in the form of written - closed / open book, individual / group, home assignment, or in the form of oral presentation, quiz, seminar etc.
6. Show complete split up of sessional marks in the page "Particulars of Marks". Final sessional mark for each student should be equal to the sum of marks awarded for Assignments (10) and Series
Tests (40).
7. All the entries in the course diary must be, legibly written without overwriting and free of errors.
8. Do not count marks of class tests along with the series test for computing sessional mark.
9. The staff member will be responsible for the safe custody of the Course Diary and (s)he should return it to the HOD at the end of semester or earlier if (s)he leaves the department or discontinue the
subject.
10. Follow KTU regulations for computing sessional marks.
PRINCIPAL
CST395 - NEURAL NETWORKS AND DEEP LEARNING
3

4. VISION OF THE COLLEGE
"A globally renowned institution of excellence in engineering, education, research and consultancy."
MISSION OF THE COLLEGE
"To contribute to the society by providing quality education and training, leading to innovation , entrepreneurship and sustainable growth."
CST395 - NEURAL NETWORKS AND DEEP LEARNING
4

5. Heading Content
PO1 Engineering knowledge: Apply the knowledge of mathematics, science, engineering fundamentals, and an engineering specialization to the solution of complex engineering problems
PO2
Problem analysis: Identify, formulate, review research literature, and analyze complex engineering problems reaching substantiated conclusions using first principles of mathematics, natural sciences, and engineering
sciences.
PO3
Design/development of solutions: Design solutions for complex engineering problems and design system components or processes that meet the specified needs with appropriate consideration for the public health and
safety, and the cultural, societal, and environmental considerations.
PO4
Conduct investigations of complex problems: Use research-based knowledge and research methods including design of experiments, analysis and interpretation of data, and synthesis of the information to provide valid
conclusions
PO5
Modern tool usage: Create, select, and apply appropriate techniques, resources, and modern engineering and IT tools including prediction and modeling to complex engineering activities with an understanding of the
limitations
PO6
The engineer and society: Apply reasoning informed by the contextual knowledge to assess societal, health, safety, legal and cultural issues and the consequent responsibilities relevant to the professional engineering
practice
PO7 Environment and sustainability: Understand the impact of the professional engineering solutions in societal and environmental contexts, and demonstrate the knowledge of, and need for sustainable development
PO8 Ethics: Apply ethical principles and commit to professional ethics and responsibilities and norms of the engineering practice
PO9 Individual and team work: Function effectively as an individual, and as a member or leader in diverse teams, and in multidisciplinary settings
PO10
Communication: Communicate effectively on complex engineering activities with the engineering community and with society at large, such as, being able to comprehend and write effective reports and design
documentation, make effective presentations, and give and receive clear instructions.
PO11
Project management and finance: Demonstrate knowledge and understanding of the engineering and management principles and apply these to one‘s own work, as a member and leader in a team, to manage projects and
in multidisciplinary environments
PO12 Life-long learning: Recognize the need for, and have the preparation and ability to engage in independent and life-long learning in the broadest context of technological change
PROGRAM OUTCOME
CST395 - NEURAL NETWORKS AND DEEP LEARNING
5

6. VISION OF COMPUTER SCIENCE AND ENGINEERING DEPARTMENT
"A centre of excellence, in the field of Computer Science and Engineering education and research, which extends its appreciated services to the industry and society."
MISSION OF COMPUTER SCIENCE AND ENGINEERING DEPARTMENT
"To develop engineers with excellent analytic, design and implementation skills, who can expertise themselves as computer professionals, research engineers, entrepreneurs or as managers, while fulfilling their
ethical and social responsibilities, in a globally competitive environment."
CST395 - NEURAL NETWORKS AND DEEP LEARNING
6

7. SlNo. Topic
1 Demonstrate the basic concepts of machine learning models and performance measures.
2 Illustrate the basic concepts of neural networks and its practical issues
3 Outline the standard regularization and optimization techniques for deep neural networks
4 Build CNN and RNN models for different use cases
5 Explain the concepts of modern RNNs like LSTM, GRU
COURSE OUTCOME
CST395 - NEURAL NETWORKS AND DEEP LEARNING
7

8. Heading Content
PEO1 Be employed as computer science professionals beyond entry-level positions or be making satisfactory progress in graduate programs.
PEO2 Be able to route their talents in to post graduate and research programs, promoting remarkable advancements in emerging areas.
PEO3
Have peer-recognized expertise together with the ability to articulate that expertise as computer science professionals.

PEO4 Apply good analytic, design, and implementation skills required to formulate and solve computer science problems.
PEO5 Demonstrate that they can function, communicate, collaborate and continue to learn effectively as ethically and socially responsible computer science professionals.
PROGRAM EDUCATIONAL OBJECTIVES
Heading Content
PSO1 Foundation of Computer System: Ability to understand the principles and working of computer systems. Students can assess the hardware and software aspects of computer systems.
PSO2
Foundations of Software development: Ability to understand the structure and development methodologies of software systems. Possess professional skills and knowledge of software design process. Familiarity and
practical competence with a broad range of programming language and open source platforms.
PSO3 Foundation of mathematical concepts: Ability to apply mathematical methodologies to solve computation task, model real world problem using appropriate data structure and suitable algorithm.
PSO4 Applications of Computing and Research Ability: Ability to use knowledge in various domains to identify research gaps and hence to provide solution to new ideas and innovations.
PROGRAM SPECIFIC OUTCOMES
CST395 - NEURAL NETWORKS AND DEEP LEARNING
8

9. Sep 2022 Oct 2022
Description Description
Sat 01
Sun 02
Mon 03
Tue 04
Wed 05
Thu 06
Fri 07
Sat 08
Sun 09
Mon 10
Tue 11
Wed 12
Thu 13
Fri 14
Sat 15
Sun 16
Mon 17
Tue 18
Wed 19
Thu 20
Fri 21
Thu 22 Sat 22
Fri 23 Sun 23
Sat 24 Mon 24
Sun 25 Tue 25
Mon 26 Wed 26
Tue 27 Thu 27
Wed 28 Fri 28
Thu 29 Sat 29
Fri 30 Sun 30
Mon 31
ACADEMIC CALENDAR
Days Date Class Days Date Class
CST395 - NEURAL NETWORKS AND DEEP LEARNING
9

10. Nov 2022 Dec 2022
Description Description
Tue 01 Thu 01
Wed 02 Fri 02
Thu 03 Sat 03
Fri 04 Sun 04
Sat 05 Mon 05
Sun 06 Tue 06
Mon 07 Wed 07
Tue 08 Thu 08
Wed 09 Fri 09
Thu 10 Sat 10
Fri 11 Sun 11
Sat 12 Mon 12
Sun 13 Tue 13
Mon 14 Wed 14
Tue 15 Thu 15
Wed 16 Fri 16
Thu 17 Sat 17
Fri 18 Sun 18
Sat 19 Mon 19
Sun 20 Tue 20
Mon 21 Wed 21
Tue 22 Thu 22
Wed 23 Series Exam 1 Fri 23
Thu 24 Sat 24
Fri 25 Sun 25
Sat 26 Mon 26
Sun 27 Tue 27
Mon 28 Wed 28
Tue 29 Thu 29
Wed 30 Fri 30
Sat 31
ACADEMIC CALENDAR
Days Date Class Days Date Class
CST395 - NEURAL NETWORKS AND DEEP LEARNING
10

11. Jan 2023
Description
Sun 01
Mon 02
Tue 03
Wed 04
Thu 05
Fri 06
ACADEMIC CALENDAR
Days Date Class
CST395 - NEURAL NETWORKS AND DEEP LEARNING
11

12. Day Period 1 Period 2 Period 3 Period 4 Period 5 Period 6 Period 7 Period 8 Period 9
Monday -- -- -- -- --
CST395 - NEURAL
NETWORKS AND
DEEP LEARNING
-- -- --
Tuesday -- -- -- -- -- -- -- -- --
Wednesday -- -- -- -- --
CST395 - NEURAL
NETWORKS AND
DEEP LEARNING
CST395 - NEURAL
NETWORKS AND
DEEP LEARNING
-- --
Thursday -- -- -- -- -- -- -- -- --
Friday -- --
CST395 - NEURAL
NETWORKS AND
DEEP LEARNING
-- -- -- -- -- --
TIMETABLE
CST395 - NEURAL NETWORKS AND DEEP LEARNING
12

13. Module Topic
1
1.1. Basics of Machine Learning
1.1.1.Machine Learning basics - Learning algorithms - Supervised, Unsupervised, Reinforcement, Overfitting, Underfitting, Hyper parameters and Validation sets
1.1.2.Estimators -Bias and Variance. Challenges in machine learning
1.1.3.Simple Linear Regression, Logistic Regression
1.1.4.Performance measures - Confusion matrix, Accuracy, Precision, Recall, Sensitivity, Specificity, Receiver Operating Characteristic curve( ROC), Area Under Curve(AUC)
2
2.1. Neural Networks
2.1.1.Introduction to neural networks
2.1.2.Single layer perceptrons, Multi Layer Perceptrons (MLPs), Representation Power of MLPs
2.1.3.Activation functions - Sigmoid, Tanh, ReLU, Softmax
2.1.4.Risk minimization, Loss function
2.1.5.Training MLPs with backpropagation
2.1.6.Practical issues in neural network training - The Problem of Overfitting, Vanishing and exploding gradient problems
2.1.7.Difficulties in convergence, Local and spurious Optima, Computational Challenges
2.1.8.Applications of neural networks
3
3.1. Deep learning
3.1.1.Introduction to deep learning, Deep feed forward network, Training deep models
3.1.2.Optimization techniques - Gradient Descent (GD), GD with momentum
3.1.3.Nesterov accelerated GD, Stochastic GD, AdaGrad, RMSProp, Adam
3.1.4.Regularization Techniques - L1 and L2 regularization, Early stopping, Dataset augmentation
3.1.5.Parameter sharing and tying, Injecting noise at input, Ensemble methods, Dropout, Parameter initialization
4
4.1. Convolutional Neural Network
4.1.1.Convolutional Neural Networks – Convolution operation, Motivation, Pooling, Convolution and Pooling as an infinitely strong prior
4.1.2.Variants of convolution functions, Structured outputs, Data types
4.1.3.Efficient convolution algorithms
4.1.4.Practical use cases for CNNs
4.1.5.Case study - Building CNN model AlexNet with handwritten digit dataset MNIST
5
5.1. Recurrent Neural Network
5.1.1.Recurrent neural networks – Computational graphs, RNN design
5.1.2.encoder – decoder sequence to sequence architectures
5.1.3.deep recurrent networks, recursive neural networks
5.1.4.modern RNNs LSTM and GRU
5.1.5.Practical use cases for RNNs
5.1.6.Case study - Natural Language Processing
SYLLABUS
CST395 - NEURAL NETWORKS AND DEEP LEARNING
13

14. Date Hour Topic Module Method Of Delivery
2022-10-19 Hour 7
Machine Learning basics - Learning algorithms - Supervised, Unsupervised, Reinforcement, Overfitting, Underfitting, Hyper parameters and
Validation sets
1 Lecture
2022-10-21 Hour 3 Estimators -Bias and Variance. Challenges in machine learning 1 Lecture
2022-10-24 Hour 6 Estimators -Bias and Variance. Challenges in machine learning 1 Unknown
2022-10-26
Hour 6 Simple Linear Regression, Logistic Regression 1 Lecture
Hour 7
Performance measures - Confusion matrix, Accuracy, Precision, Recall, Sensitivity, Specificity, Receiver Operating Characteristic curve( ROC),
Area Under Curve(AUC)
1 Lecture
2022-10-28 Hour 3
Performance measures - Confusion matrix, Accuracy, Precision, Recall, Sensitivity, Specificity, Receiver Operating Characteristic curve( ROC),
Area Under Curve(AUC)
1 Unknown
2022-10-31 Hour 6 Introduction to neural networks 2 Lecture
2022-11-02
Hour 6 Single layer perceptrons, Multi Layer Perceptrons (MLPs), Representation Power of MLPs 2 Lecture
Hour 7 Single layer perceptrons, Multi Layer Perceptrons (MLPs), Representation Power of MLPs 2 Unknown
2022-11-04 Hour 3 Activation functions - Sigmoid, Tanh, ReLU, Softmax 2 Lecture
2022-11-07 Hour 6 Risk minimization, Loss function 2 Lecture
2022-11-09
Hour 6 Training MLPs with backpropagation 2 Unknown
Hour 7 Training MLPs with backpropagation 2 Lecture
2022-11-11 Hour 3 Practical issues in neural network training - The Problem of Overfitting, Vanishing and exploding gradient problems 2 Lecture
2022-11-14 Hour 6 Training MLPs with backpropagation 2 Lecture
2022-11-16
Hour 6 Difficulties in convergence, Local and spurious Optima, Computational Challenges 2 Lecture
Hour 7 Applications of neural networks 2 Lecture
2022-11-21 Hour 6 Introduction to deep learning, Deep feed forward network, Training deep models 3 Lecture
2022-11-23 Hour 6 Optimization techniques - Gradient Descent (GD), GD with momentum 3 Lecture
2022-11-28 Hour 6 Nesterov accelerated GD, Stochastic GD, AdaGrad, RMSProp, Adam 3 Lecture
2022-11-30 Hour 6 Regularization Techniques - L1 and L2 regularization, Early stopping, Dataset augmentation 3 Lecture
2022-12-02 Hour 3 Parameter sharing and tying, Injecting noise at input, Ensemble methods, Dropout, Parameter initialization 3 Lecture
2022-12-05 Hour 6 Convolutional Neural Networks – Convolution operation, Motivation, Pooling, Convolution and Pooling as an infinitely strong prior 4 Lecture
2022-12-07
Hour 6 Variants of convolution functions, Structured outputs, Data types 4 Lecture
Hour 7 Efficient convolution algorithms 4 Lecture
2022-12-12 Hour 6 Practical use cases for CNNs 4 Lecture
2022-12-14 Hour 6 Case study - Building CNN model AlexNet with handwritten digit dataset MNIST 4 Lecture
2022-12-16 Hour 3 Recurrent neural networks – Computational graphs, RNN design 5 Lecture
2022-12-19 Hour 6 encoder – decoder sequence to sequence architectures 5 Lecture
2022-12-21 Hour 7 deep recurrent networks, recursive neural networks 5 Lecture
SUBJECT PLAN
CST395 - NEURAL NETWORKS AND DEEP LEARNING
14

15. 2022-12-23 Hour 3 modern RNNs LSTM and GRU 5 Lecture
2023-01-04
Hour 6 Practical use cases for RNNs 5 Lecture
Hour 7 Case study - Natural Language Processing 5 Lecture
Date Hour Topic Module Method Of Delivery
CST395 - NEURAL NETWORKS AND DEEP LEARNING
15

16. Date Hour Topic Module Remarks
2022-10-14 Hour 3
Machine Learning basics - Learning algorithms - Supervised, Unsupervised, Reinforcement, Overfitting, Underfitting, Hyper
parameters and Validation sets
1
2022-10-17 Hour 6
Machine Learning basics - Learning algorithms - Supervised, Unsupervised, Reinforcement, Overfitting, Underfitting, Hyper
parameters and Validation sets
1
2022-10-19
Hour 6 Estimators -Bias and Variance. Challenges in machine learning 1
Hour 7 Simple Linear Regression, Logistic Regression 1
2022-10-21 Hour 3 Simple Linear Regression, Logistic Regression 1
2022-10-24 Hour 6
Performance measures - Confusion matrix, Accuracy, Precision, Recall, Sensitivity, Specificity, Receiver Operating
Characteristic curve( ROC), Area Under Curve(AUC)
1
2022-10-26
Hour 6
Performance measures - Confusion matrix, Accuracy, Precision, Recall, Sensitivity, Specificity, Receiver Operating
Characteristic curve( ROC), Area Under Curve(AUC)
1
Hour 7
Performance measures - Confusion matrix, Accuracy, Precision, Recall, Sensitivity, Specificity, Receiver Operating
Characteristic curve( ROC), Area Under Curve(AUC)
1
2022-10-28 Hour 3 Practical use cases for CNNs
4
2022-10-31 Hour 6
Performance measures - Confusion matrix, Accuracy, Precision, Recall, Sensitivity, Specificity, Receiver Operating
Characteristic curve( ROC), Area Under Curve(AUC)
1
2022-11-02
Hour 6 Training MLPs with backpropagation 2
Hour 7 Introduction to neural networks 2
2022-11-04 Hour 3 Single layer perceptrons, Multi Layer Perceptrons (MLPs), Representation Power of MLPs 2
2022-11-11 Hour 3 Activation functions - Sigmoid, Tanh, ReLU, Softmax 2
2022-11-14 Hour 6
Training MLPs with backpropagation 2
Risk minimization, Loss function 2
2022-11-16
Hour 6 Practical issues in neural network training - The Problem of Overfitting, Vanishing and exploding gradient problems 2
Hour 7
Difficulties in convergence, Local and spurious Optima, Computational Challenges 2
Applications of neural networks 2
2022-11-18 Hour 3 Introduction to deep learning, Deep feed forward network, Training deep models 3
2022-11-21 Hour 6
Convolutional Neural Networks – Convolution operation, Motivation, Pooling, Convolution and Pooling as an infinitely
strong prior
4
SUBJECT COVERAGE
CST395 - NEURAL NETWORKS AND DEEP LEARNING
16

17. S.NoName Type Date Max Mark Result Status Question
1 Series Exam 1 OFFLINE EXAM 23/11/2022 50 Results not published
SERIES EXAMS
CST395 - NEURAL NETWORKS AND DEEP LEARNING
17

18. Total
Roll No Name Grace Marks
Final Score
(out of 50)
1 ABDU NAFIH V P 0 0
4 AJMILA SHADA 0 0
8 AMAL GANESH 0 0
19 ARJUN P JAYAKRISHNAN 0 0
22 ASWANTH K 0 0
26
FAIZAN THEKKE PULUKKOOL
0 0
37 MUHAMMAD ASAD V P 0 0
47 SANDEEP E P 0 0
48 SAYOOJ SATHYAN 0 0
52 TONA ANTONY 0 0
55 VISMAYA B 0 0
64 RAMNATH SHENOY B 0 0
65 ATHUL BENNY 0 0
INTERNAL MARK
CST395 - NEURAL NETWORKS AND DEEP LEARNING
18

19. Course Exit Survey Details Not Found
COURSE EXIT SURVEY
CST395 - NEURAL NETWORKS AND DEEP LEARNING
19

20. S.NoDate Event Remarks
YEAR CALENDAR
CST395 - NEURAL NETWORKS AND DEEP LEARNING
20

21. Roll No Student
14/1017/1019/1021/1024/1026/1028/1031/1002/1104/1111/1114/1116/1118/1121/11
Total
3 6 67 3 6 67 3 6 67 3 3 6 67 3 6
1 ABDU NAFIH V P P P P P P P P P P P P P P P P P P P P 19/19 (100)
4 AJMILA SHADA P P P P P P P P P P P P P P P P P P P 19/19 (100)
8 AMAL GANESH P P P P P P P P P P P P P P P P P P P 19/19 (100)
19 ARJUN P JAYAKRISHNAN P P P P P P P P P P P P P P P P P P P 19/19 (100)
22 ASWANTH K P P P P P P P P P P P P P P P P P P P 19/19 (100)
26 FAIZAN THEKKE PULUKKOOL P P P P P P P P P P P P P P P P P P P 19/19 (100)
37 MUHAMMAD ASAD V P P P P P P P P P P P P P P P P P P P P 19/19 (100)
47 SANDEEP E P P P P P P P P P P P P P P P P P P P P 19/19 (100)
48 SAYOOJ SATHYAN P P P P P P P P P P P P P P P P P P P 19/19 (100)
52 TONA ANTONY P P P P P P P P P P P P P P P P P P P 19/19 (100)
55 VISMAYA B P P P P P P P P P P P P P P P P P P P 19/19 (100)
64 RAMNATH SHENOY B P P P P P P P P P P P P P P P P P P P 19/19 (100)
65 ATHUL BENNY P P P P P P P P P P P P P P P P P P P 19/19 (100)
ATTENDANCE SUMMARY
CST395 - NEURAL NETWORKS AND DEEP LEARNING
21

22. Roll No Student Total
1 ABDU NAFIH V P 19/19 (100)
4 AJMILA SHADA 19/19 (100)
8 AMAL GANESH 19/19 (100)
19 ARJUN P JAYAKRISHNAN 19/19 (100)
22 ASWANTH K 19/19 (100)
26 FAIZAN THEKKE PULUKKOOL 19/19 (100)
37 MUHAMMAD ASAD V P 19/19 (100)
47 SANDEEP E P 19/19 (100)
48 SAYOOJ SATHYAN 19/19 (100)
52 TONA ANTONY 19/19 (100)
55 VISMAYA B 19/19 (100)
64 RAMNATH SHENOY B 19/19 (100)
65 ATHUL BENNY 19/19 (100)
ATTENDANCE SUMMARY
CST395 - NEURAL NETWORKS AND DEEP LEARNING
22