```
Introduction to Machine Learning cover Dark cover graphic with scatter plot clusters, a decision tree, and a neural network motif surrounding a title panel n Σ i = 1 y = Xβ + ε σ(x) min θ ŷ = f(x; θ) ∇L Introduction to Machine Learning Supplementary Course Notes Sajjad Haider Clustering Decision Trees Neural Networks Regression Classification

Preface

Welcome to the Fall 2026 semester. These notes have been prepared to support the lectures, assignments, and other course activities. The objective is to provide a single place where you can review the concepts discussed in class, work through numerical examples, and practice the implementation of different machine learning techniques.

The material is presented with a combination of explanations, examples, numerical solutions, code, and interactive elements. The purpose of these interactive components is to make it easier to understand the connection between the concepts discussed in class and their implementation. You are encouraged to use the examples and exercises while studying rather than simply reading through the material.

The course provides an introduction to the main concepts and techniques used in machine learning, with an emphasis on understanding the underlying ideas as well as their practical implementation using Python and commonly used libraries. The topics covered in the course include:

  • Classification: k-nearest neighbors, classification trees, naïve Bayes, logistic regression, neural networks, and ensembles (Random Forest, AdaBoost, and Stacking).
  • Regression: Ordinary least squares regression, regression trees, and gradient boosting.
  • Clustering: Agglomerative clustering, K-means, and DBSCAN.
  • Dimensionality Reduction & Feature Selection: Principal Component Analysis (PCA), auto-encoders, recursive feature elimination, and forward/backward/step-wise selection.

In addition to these algorithms, the course covers model evaluation and validation, including train/test/validation splits, cross-validation, and performance measures for classification and regression. We will also discuss automated machine learning (AutoML), model selection, and hyperparameter optimization. MLOps concepts related to the machine learning lifecycle will also be introduced through a self-paced certification module.

The contents of these notes are not meant to replace the lectures or the recommended textbooks. They should be used as a supplementary resource for the course. I would also encourage you to work through the examples yourself, try the code, and attempt the exercises and quizzes. This will help you understand the concepts better than simply reading the material.

A Note About These Notes

Some of the material in these notes has been prepared with the assistance of AI-based tools, particularly for organizing content, preparing examples, formatting, and generating some diagrams and interactive components. The material has been reviewed and adapted for use in this course, but AI-generated content can still contain errors or inaccuracies.

The contents will therefore be revised as the course progresses. Some sections may be modified, expanded, or removed depending on the topics covered in lectures. Corrections will also be made if errors or inconsistencies are identified. Students are encouraged to report any such issues so that they can be corrected in subsequent versions.

Reference Books

The following books and resources have been used in preparing and organizing the course material. Students are encouraged to refer to them for additional explanations and examples.

  • Machine Learning with PyTorch and Scikit-Learn by Raschka, Liu, and Mirjalili (2022)
    GitHub Repository
  • Grokking Machine Learning by Luis Serrano (2021)
    Manning Publications
  • An Introduction to Statistical Learning – With Applications in Python (2024 Edition)
    Official Resources
```