Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 35 views

    SML Syllabus

    Uploaded by

    anmolsupr.official
    This document provides a syllabus for a Statistical Machine Learning course. The course will cover key machine learning concepts including supervised learning, unsupervised learning, reinforcement learning, linear regression, logistic regression, decision trees, ensemble methods, neural networks, dimensionality reduction, and clustering. It is a 4 credit course taught over 3 lecture hours and 2 lab hours per week. Students will implement machine learning algorithms and evaluate models in lab sessions. The course aims to give students skills in formulating problems as statistical machine learning tasks and implementing common techniques.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    SML Syllabus

    Uploaded by

    anmolsupr.official
    35 views3 pages
    This document provides a syllabus for a Statistical Machine Learning course. The course will cover key machine learning concepts including supervised learning, unsupervised learning, reinforcement learning, linear regression, logistic regression, decision trees, ensemble methods, neural networks, dimensionality reduction, and clustering. It is a 4 credit course taught over 3 lecture hours and 2 lab hours per week. Students will implement machine learning algorithms and evaluate models in lab sessions. The course aims to give students skills in formulating problems as statistical machine learning tasks and implementing common techniques.

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document provides a syllabus for a Statistical Machine Learning course. The course will cover key machine learning concepts including supervised learning, unsupervised learning, reinforcement learning, linear regression, logistic regression, decision trees, ensemble methods, neural networks, dimensionality reduction, and clustering. It is a 4 credit course taught over 3 lecture hours and 2 lab hours per week. Students will implement machine learning algorithms and evaluate models in lab sessions. The course aims to give students skills in formulating problems as statistical machine learning tasks and implementing common techniques.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    35 views3 pages

    SML Syllabus

    Uploaded by

    anmolsupr.official
    This document provides a syllabus for a Statistical Machine Learning course. The course will cover key machine learning concepts including supervised learning, unsupervised learning, reinforcement learning, linear regression, logistic regression, decision trees, ensemble methods, neural networks, dimensionality reduction, and clustering. It is a 4 credit course taught over 3 lecture hours and 2 lab hours per week. Students will implement machine learning algorithms and evaluate models in lab sessions. The course aims to give students skills in formulating problems as statistical machine learning tasks and implementing common techniques.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 3

    Statistical Machine Learning Syllabus

    COURSE CONTEXT

    COURSE CONTEXT
    SCHOOL SCSET VERSION NO. OF V1
    CURRICULUM/SYLLABUS
    THAT THIS COURSE IS A PART
    OF
    DEPARTMENT DATE THIS COURSE WILL BE July-Dec,
    EFFECTIVE FROM 2023
    DEGREE B. Tech. VERSION NUMBER OF THIS 2
    COURSE

    COURSE BRIEF
    COURSE TITLE Statistical Machine Learning PRE-REQUISITES NA
    COURSE CODE CSET211 TOTAL CREDITS 4
    COURSE TYPE Specialized Core-I L-T-P FORMAT 3-0-2

    COURSE SUMMARY
    This course includes machine learning concepts, Statistical Theories, Supervised learning, high dimensional data
    and the role of sparsity, Learning theory, Risk minimization, Classification and regression, and EM algorithm. It
    also covers important topics such as parametric and non-parametric methods, theory of generalization,
    regularization, the role of sparsity in high dimensional data, and surrogate loss functions. In a broader sense, the
    course offers a thorough understanding of statistical ML concepts that help students design and implement daily
    life learning applications.

    COURSE-SPECIFIC LEARNING OUTCOMES (CO)


    By the end of this program, students should have the following knowledge, skills and values:
    CO1: To articulate key features and methods of Statistical Machine Learning (SML).
    CO2: To formulate and design the given application as a statistical machine learning problem.
    CO3: To implement and evaluate common statistical machine learning techniques.

    How are the above COs aligned with the Program-Specific Objectives (POs) of the degree?
    The course outcomes are aligned to inculcating inquisitiveness in understanding cutting edge areas of
    computer science engineering and allied disciplines with their potential impacts.

    CO - PO Mapping

    COs  POs PO1 PO2 PO3 PO4 PO5 PO6


    CO1

    CO2

    CO3

    Detailed Syllabus
    Module 1 (8 hours)
    Designing a learning system, Types of machine learning: Problem based learning, Supervised learning,
    Unsupervised learning, Reinforcement learning, Linear Regression: Weights and Features,
    Applications, Cost Functions, Finding best fit line, Gradient Descent Algorithm: Learning Algorithm,
    First order derivatives, Linear regression using gradient descent, Learning theory, Risk minimization,
    Learning rate, Logistic Regression, Sigmoid Function, Cost Function for Logistic Regression, Multi-
    class classification, Probability Distribution, Polynomial Regression, Performance Metrics:
    Classification (Confusion Matrix, Accuracy, Precision, Recall, F1-score, ROC-AUC), Regression
    (MSE, MAE, RMSE, R2 Score).

    Module 2 (10 hours)


    Decision Tree, Selecting Best Splitting Attribute, CART (Gini Index). ID3 (Entropy, Information
    Gain), Hyperparameters in Decision tree, Issues in Decision tree learning. Overfitting and
    Underfitting, Bias and Variance, Theory of generalization, regularization, Cross Validation. Ensemble
    Learning (Concord’s Theorem), Bagging, Bootstrap and Aggregation, Random Forest. Boosting,
    AdaBoost, Gradient Boost. Feature Engineering, Feature Selection, Feature Extraction, Convexity and
    Optimization, PAC learning

    Module 3 (12 hours)


    Artificial Neural Network, Neural network representation, Perceptron model, Stepwise v/s Sigmoid
    function, Multilayer perceptron model, Matrix Calculus (Jacobian, Hessian Matrix), Computation
    Graph, Backpropagation Algorithm, Activation Functions, Stochastic Gradient Descent, Batch
    Gradient Descent, Mini-Batch Gradient Descent, Vanishing and Exploding Gradients, Overfitting
    Problem, Sparsity, Regularization (Ridge, Lasso, Elastic), Dropout and Early Stopping, Bayesian
    Learning: Bayes theorem and concept learning, Naïve Bayes classifier, Gibbs Algorithm, Bayesian
    belief networks, The EM algorithm, Gaussian Distributions, Gaussian Mixture Models, Gaussian
    Discriminant Analysis, Support Vector Machines, Hyperplane, Support Vectors, Kernels, Non-
    Parametric Regression, Locally weighted regression, K-nearest neighbour.

    Module 4 (12 hours)


    Unsupervised learning (clustering, Association rule learning, Dimensionality reduction), Common
    distance Measures, k-means clustering, Elbow method, Hierarchical Clustering – agglomerative and
    divisive, Dendogram, Similarity measures for hierarchical clustering, DBSCAN, Cluster Quality (R
    index, Silhouette Coefficient), Dimensionality Reduction, Principal Component Analysis, Singular
    Vector Decomposition, T-distributed Stochastic Neighbour Embedding.

    STUDIO WORK / LABORATORY EXPERIMENTS:


    Students will gain practical experience with the implementation of different statistical methods by using different
    statistical machine learning tools. Eventually, the lab works formulate the problem as a statistical machine learning
    problem followed by its implementation.

    TEXTBOOKS/LEARNING RESOURCES:
    a) Masashi Sugiyama, Introduction to Statistical Machine Learning (1 ed.), Morgan Kaufmann, 2017.
    ISBN 978-0128021217.
    b) T. M. Mitchell, Machine Learning (1 ed.), McGraw Hill, 2017. ISBN 978-1259096952.

    REFERENCE BOOKS/LEARNING RESOURCES:


    a) Richard Golden, Statistical Machine Learning A Unified Framework (1 ed.), CRC Press, 2020. ISBN
    9781351051490.

    TEACHING-LEARNING STRATEGIES
    The course will be taught using a combination of the best practices of teaching-learning. Multiple environments
    will be used to enhance the outcomes such as seminar, self-learning, MOOCs, group discussions and ICT based
    tools for class participation along with the classroom sessions. The teaching pedagogy being followed includes
    more exposure to hands-on experiment and practical implementations done in the lab sessions. To match with the
    latest trend in academics, case study, advanced topics and research oriented topics are covered to lay down the
    foundation and develop the interest in the students leading to further exploration of the related topics. To make the
    students aware of the industry trends, one session of expert lecture will be organized to provide a platform to the
    students for understanding the relevant industry needs.
    EVALUATION POLICY

    Components of Course Evaluation Percentage Distribution

    Midterm 15
    End Semester Examination 35
    Continuous Lab Evaluation 30
    Quiz 5
    Certification 15
    Total 100

    You might also like