This repository contains a comprehensive set of Machine Learning lecture notes, written in LaTeX, together with supporting code, figures, and project reports.
The material is designed to be rigorous, mathematically grounded, and suitable for university-level study.
The core of this repository is the LaTeX manuscript:
Machine Learning.tex
which compiles into:
Machine Learning.pdf
The notes cover the theoretical and practical foundations of Machine Learning, including:
- Introduction to Machine Learning
- Linear Models
- Regression
- Classification
- Model Evaluation
- Neural Networks
- Decision Trees
- Random Forests
- Support Vector Machines
- Ensemble Learning
- Theoretical Insights and Seminar Extensions
All chapters are organized in the chapters/ directory and structured modularly for clarity and maintainability.
The document aims to:
- Provide formal derivations and mathematical rigor
- Connect theory and implementation
- Include illustrative figures and diagrams
- Offer intuitive explanations alongside proofs
- Present complete worked examples
Below is the logical organization of the repository :
Contains all main LaTeX chapter sources.
Contains front matter material (e.g., license page).
Contains LaTeX source files for project reports.
Python scripts and experimental code supporting the notes.
Example:
adaboost_decision_boundary.py
Vector drawings (.drawio) used to generate high-quality diagrams.
All figures used in the notes (plots, diagrams, illustrations, schemas).
This section contains projects developed and discussed in the notes.
Description: This project focuses on developing and evaluating machine learning models for medical MRI diagnostics, specifically aiming to classify and interpret MRI scans for diagnostic purposes. The work explores deep learning architectures and training strategies to improve automated analysis of MRI images.
Objective: The main learning task is medical image classification, where the goal is to train a model capable of accurately identifying pathological features in MRI scans. This involves supervised learning using labeled medical imaging datasets.
Files:
projects/mri-classification.tex- Associated dataset directories and preprocessing scripts within the linked repository
The project uses a Makefile to manage compilation.
All auxiliary files are generated inside the out/ directory, and the final PDF is copied to the root of the repository.
From the root of the repository, run:
makeThis will:
- Compile the document using
latexmk - Automatically run BibTeX if required
- Store all intermediate files (
.aux,.log,.toc, etc.) insideout/ - Copy the final
Machine Learning.pdfto the repository root
To remove all generated auxiliary files:
make clean- A LaTeX distribution (TeX Live, MiKTeX, etc.)
latexmkbibtex
Contributions are very welcome.
You can contribute by:
- Opening issues (typos, conceptual mistakes, improvements)
- Submitting pull requests
If you submit a pull request, please ensure that:
- ✅ You modify the LaTeX source files, not only the PDF
- ✅ You confirm that the project compiles successfully using method in section above
- ✅ You upload the newly compiled
Machine Learning.pdf - ❗ You do not modify the repository structure unless strictly necessary
- ❗ You do not change the used libraries in Machine Learning.tex unless necessary
- ✅ You add your name to the Contributors section at the end of this README
Pull requests that do not meet these requirements may not be accepted.
-
Paolo Volpini Added notes on the MLP case study.
-
Diego Martinez Reported errors in issues #1, #2, and #3.
See frontmatter/license.tex for license details.