X

Machine Learning and Pattern Recognition

By Dr. Malaya Kumar Nath   |   NIT Puducherry
Learners enrolled: 1768
This course is a foundational course designed for UG/PG/researchers working in the area of speech recognition, computer vision, and pattern classification. It will equip learners with both theoretical foundations and practical skills required to develop intelligent systems that can learn from data and recognize meaningful patterns. This course covers a range of new and classical findings in statistical pattern categorisation and machine learning. Pattern Recognition topics such as feature extraction, decision theory, Bayesian classifiers, and discriminative models (like support vector machines and neural networks) are integrated with machine learning principles to give learners a unified understanding of intelligent data analysis. This course follows the fundamentals of machine learning techniques to demonstrate how the various tools are created, related with one another, and implemented in practice. This course avoids the potential pitfalls of merely presenting a collection of machine learning tools as though they were an end in themselves. Learners will investigate supervised and unsupervised learning techniques, including regression, classification, clustering, dimensionality reduction, and ensemble methods. In addition, the course will allow for the exploration of statistical learning theories, optimization methods, and evaluation metrics that form the backbone of modern learning algorithms.
Summary
Course Status : Upcoming
Course Type : Elective
Language for course content : English
Duration : 12 weeks
Category :
  • Teacher Education
Credit Points : 4
Level : Undergraduate/Postgraduate
Start Date : 21 Jul 2025
End Date : 30 Nov 2025
Enrollment Ends : 31 Aug 2025
Exam Date : 11 Dec 2025 IST
Translation Languages : English
NCrF Level   : 4.5
Industry Details : Teaching
Exam Shift :

Shift 2

Note: This exam date is subject to change based on seat availability. You can check final exam date on your hall ticket.


Contact NC Support


Page Visits



Course layout

Week-1: Representation of shapes and features

Ø  Introduction to pattern recognition

Ø   Shape description, Chain code, Moments, Convex hull, Medial axis

Ø   Texture, GLCM

Week-2: Linear algebra, probability, and random variables

Ø  Vectors, Distance function, Matrix operation, Tensor product

Ø  Type of matrix, Matrix calculus, Gradient, Jacobian, Hessian

Ø  Numerical optimization, Gradient descent, Newton’s method

Ø  Dilemmas in machine learning, Curse of dimensionality

Ø  Random variable, Statistical averages, Moments, Characteristic function, Cumulants

Ø  Random vectors, Statistical descriptions, Linear transformation, Real valued normal random vector, Central limit theorem

Week-3: Regression & Classification

Ø  Regression, Linear regression, Design matrix, Squared error, Moore-Penrose matrix

Ø  Linear basic function models, Logarithmic curve, Polynomial regression

Ø  Bayesian regression, Maximizing Likelihood, Bayesian learning, MAP

Ø  Classification, Bayes theorem, Naïve Bayes classifier, Decision region, Decision boundary, Error probability, Minimizing risk

Week-4: Performance evaluation

Ø  Confusion matrix, Accuracy, Precision, Recall, F1-score, Specificity, Sensitivity, Balanced accuracy, MCC, Cohen’s Kappa score, Geometric mean, Dice similarity coefficient, Conformity coefficient

Ø  Graphical methods, ROC, AUC, DET, PR curve, tSNE

Ø  Multiclass classification measure, Macro average, Micro average

Ø  MSE, UQI, SSIM, IoU, PCC, VOE

Week-5,6: Clustering

Ø  Types of clusters, Criteria of grouping elements, Distance measure, Evaluation criteria of clustering

Ø  Hierarchical clustering, K-means algorithm, K-medoid

Ø  Graph based clustering, MST, Kruskal’s algorithm, Prim’s algorithm

Ø  DBSCAN

Ø  Soft clustering, Fuzzy clustering, Overlapping K-means clustering

Week-7,8: Perceptron

Ø  Rosenblatt’s perceptron, Perceptron convergence theorem, Relation b/w perceptron and Bayes classifier, Illustration with AND & OR gate, Limitations

Ø  Multilayer perceptron, Batch learning and online learning, Back propagation algorithm, Activation functions, Rate of learning, momentum, Cross validation

Ø  XOR problem, Effective use of back propagation, Back propagation and differentiation, Hessian and its effect in online learning, Adaptive control of learning rate

Ø  Generalization, Convolutional network

Week-9: SVM

Ø  Hyperplane for linearly separable and non-linearly separable patterns, Optimum design of SVM, Examples of SVM, SVM as linear regression, XOR problem, SVM dual formulation

Week-10: Deep learning and Autoencoder

Ø  Evolution of ANN, Difference b/w AI, ML, and DL, Types of DL models, CNN, Components of CNN, Parameter computation in CNN, Pretrained models, VGG16, AlexNet

Ø  Autoencoder, Deep autoencoder, Variational autoencoder

Week-11: RBF & Dimensionality reduction

Ø  Cover’s theorem, RBF, Hybrid learning for RBF network

Ø  Principle of self-organization, Principal component analysis, Dimensionality reduction

Week-12: Information theoretic learning: Entropy, Mutual information, KL divergence, ICA

                Stochastic learning: Markov chain, Gibb’s sampling, Boltzmann machine, Deep belief net

                Dynamic programming: Markov decision process, Dynamic programming, Q-leaning

 

Assignments – 30 %, Final exam – 70 %

Books and references

1.       Haykin S., “Neural Networks and Learning Machines”, Prentice Hall, ISBN: 9780131471399

2.       C.M. Bishop, “Pattern Recognition and Machine Learning”, Springer 2006.

3.       Tom Mitchell, “Machine Learning”, McGraw Hill, ISBN:9780071154673

4.       Trevor Hastie, Robert Tibshirani, and Jerome H. Friedman, “The elements of statistical learning”, 2nd Edition, Springer 2009.

5.       Duda Hart and Stock, “Pattern Classification”, Wiley 2001.


Instructor bio

Dr. Malaya Kumar Nath

NIT Puducherry
Dr. Nath, currently working as an Assistant Professor in Department of ECE at NIT Puducherry since 2013. He is currently holding the post of Associate Dean of International Relations, Alumni, & Publicity and has served as the HoD of ECE from 05.02.2024 to 23.01.2025. He had an opportunity to work as Visiting Assistant Professor at AIT, Bangkok, in 2018. He has completed Ph.D. and M. Tech in Signal Processing from IIT Guwahati. His research focuses on machine learning, biomedical signals, & image processing. He has published 36 articles in SCI journals and 27 conference papers, with 1875 citations and h-index of 23 on Google Scholar. He is an Associate Editor for the Multimedia Tools and Applications (Springer) journal. He has supervised four Ph.D. and five M. Tech. He has organized numerous academic events, including SERB-sponsored Karyashala workshops & Symposia, GIAN Program, an AICTE-sponsored ATAL FDP. He is the senior member of IEEE.

Course certificate

"The SWAYAM Course Enrolment and learning is free. However, to obtain a certificate, the learner must register and take the proctored exam in person at one of the designated exam centres. The registration URL will be announced by NTA once the registration form becomes available. To receive the certification, you need to complete the online registration form and pay the examination fee. Additional details, including any updates, will be provided upon the publication of the exam registration form. For more information about the exam locations and the terms associated with completing the form, please refer to the form itself."

 

Grading Policy:

 

- Internal Assignment Score: This accounts for 30% of the final grade and is calculated based on the average of the best three assignments out of all the assignments given in the course.

- Final Proctored Exam Score: This makes up 70% of the final grade and is derived from the proctored exam score out of 100.

- Final Score: The final score is the sum of the average assignment score and the exam score.

 

Eligibility for Certification:

 

- To qualify for a certificate, you must achieve an average assignment score of at least 10 out of 30, and an exam score of at least 30 out of 70. If one of the 2 criteria is not met, you will not get the certificate even if the Final score >=40/100.

Certificate Details:

 

- The certificate will include your name, photograph, roll number, and the percentage score from the final exam. It will also feature the logos of the Ministry of Education, SWAYAM, and NITTTR.

- Certificate Format: Only electronic certificates (e-certificates) will be issued; hard copies will not be dispatched.

 

Once again, thanks for your interest in our online courses and certification. Happy Learning.

 

*********



MHRD logo Swayam logo

DOWNLOAD APP

Goto google play store

FOLLOW US