The Mathematics Behind Machine Learning and Artificial Intelligence

Note: This post reflects the journey of creating my first serious academic-style research paper, completed during 10th grade.

At my school, every student between 9th and 12th grade is required to complete a yearly homework project. While many chose history, literature, or biology, I wanted mine to be in mathematics—my favorite subject. At the same time, I was becoming increasingly interested in artificial intelligence, so I thought: “Why not combine both?”

At first, I wasn’t sure how deep I could go. I didn’t know if high school-level math would be enough to say something meaningful about AI. But I quickly discovered that AI systems rely on incredibly rich mathematical foundations: linear algebra for model representation, calculus for optimization, probability for decision-making, and so much more.

Over the course of several months, I read academic papers, watched lectures, and gradually constructed my own understanding of how math powers AI. The result was this paper—a structured explanation of how each mathematical discipline plays a role in making intelligent machines possible.

The process wasn’t easy. I rewrote entire sections multiple times, learning as I went. Concepts like the Universal Approximation Theorem or the Kullback-Leibler Divergence weren’t things I had seen in school. But I broke them down, step by step, until I could explain them clearly and confidently.

Below, you can find a preview version of the paper that includes the abstract and table of contents. If you'd like to read the full version, feel free to contact me.

Contact: guzmus.27@robcol.k12.tr

Preview PDF

🔗 View this project on GitHub: github.com/mustafa1guzel/mustafaguzelmathyhp