Your Machine Learning Library

Your starting point for all your machine learning needs

I get this question all the time:

“Tivadar… Where is the [Insert Topic] post?“

For long enough, my solution was to manually look up the posts, and share them with anyone who asked. And even though that works, it’s not ideal.

So, I spent a whole day going through every single post in our library and tagged each one to make them easier for you to find.

Now, instead of digging, you can just click.

The Palindrome Library

Bookmark this page to instantly access all our resources for all your math and machine learning goals:

• Graph theory,

• The fundamentals of math,

• Neural networks from scratch,

• The math of machine learning,

• The fundamentals of machine learning,

• Epsilons, or essential math concepts in 500 words or less,

• and other resources on the history (and curiosities) of mathematics.

Click any post on any category below to find resources tailored to your needs. If there’s something missing, let me know—I’m constantly expanding the library.

(Note: Not every section is complete (yet). I’m filling in the gaps over time.)

---

The Fundamentals of Math

Mathematics is a foundational tool for every machine learning practitioner. Here are the fundamentals you shouldn’t skip:

• Part 1: What is mathematics?

• Part 2: Definitions, theorems, and proofs

• Part 3: At the roots of mathematics

• Part 4: The (art and) science of reasoning

• Part 5: Proofs are (not) magic

• Part 6: Proof by induction and contradiction

• Part 7: Sets, relations, and functions

---

Mathematics of Machine Learning

These are the core topics of mathematics you need for Machine Learning. I narrowed them down to three main categories:

Linear Algebra

• Matrices and graphs

• How to measure the angle between two functions

• The non-recursive formula for Fibonacci numbers

• The unreasonable effectiveness of orthogonal systems

Probability Theory:

• The science of science

• The law of large numbers

• 1111111111 is a random sequence

• Your brain is wrong about probability

• Is probability frequentist or Bayesian?

• How large is the law of large numbers?

• Probabilities, densities, and distributions

• What’s the meaning of the expected value

Calculus:

• Why does gradient descent work

• The history of trigonometric functions

• How to add infinitely many numbers together

• How the Babylonians knew √2 up to six digits

• The fascinating story of the exponential function

Here’s a more comprehensive roadmap of the math you need for machine learning:

• The Roadmap of Mathematics for Machine Learning

---

Epsilons

A series where I explain mathematical concepts or tools in less than 500 words:

• Part 1: The geometric series

• Part 2: Matrix multiplication

• Part 3: The LU decomposition

• Part 4: The Gram-Schmidt process

• Part 5: The QR decomposition

• Part 6: What is that Σ?

• Part 7: What is a kernel?

---

Graph Theory

Graphs are one of the most useful abstractions in math and computer science. Here’s our introductory series to help you understand and use them:

• Part 1: What your brain, your friends, and your city have in common

• Part 2: Trees, forests, and spanning subgraphs

• Part 3: An introduction to matchings

• Part 4: Eulerian and Hamiltonian tours

• Part 5: Graphs and their drawings

---

The Fundamentals of Machine Learning

PyTorch, TensorFlow, scikit-learn: they revolutionized machine learning, enabling rapid prototyping and development. But if you want to be a machine learning engineer, you have to understand what’s underneath the hood.

These articles will help you familiarize yourself with every nook and cranny:

The theory:

• Vectorization in theory

• Vectorization in practice

• Why does gradient descent work

• Machine learning is not just statistics

• Why are neural networks so powerful?

• The taxonomy of machine learning paradigms

• Where does the mean squared error come from?

• The mathematics of optimization for deep learning

• What Full Body Composition Scales Teach About Machine Learning

The algorithms:

• Logistic regression

• How to grow a decision tree

• How to implement a decision tree

• The Anatomy of Logistic Regression

• Understanding k-Nearest Neighbors

• Linear regression in multiple variables

• Building an interface for machine learning

• Linear regression and the least squares problem

• Linear Regression and The Exploding Gradients

• Linear regression (is the most important algorithm ever)

---

Neural Networks from Scratch

Neural networks are a fundamental part of modern machine learning. This series teaches you how to build one from scratch:

• Part 1: Introduction to Computational Graphs

• Part 2: Computational Graphs as Neural Networks

• Part 3: Computational Graphs and the Forward Pass

• Part 4: The Derivatives of Computational Graphs

• Part 5: Forward vs Backward Differentiation

---

Algorithmic Analysis

A series on how to think about the correctness, cost, and performance of algorithms.

• Part 1: Introduction to Algorithmic Analysis

---

Miscelaneous

Articles on the history of mathematics and how it shapes culture, life, and work:

• What is infinity?

• The camel principle

• The obstacle is the path

• Visualizing mathematics

• Education or experience?

• A conversation about education

• There are many roads to machine learning

• How does the Japanese multiplication work

• How the Babylonians knew √2 up to six digits

• What Elden Ring taught me about hard things

• The most important skill you were never taught

• Do you have to learn math for machine learning?

• Can Large Language Models replace mathematicians?

• What full body composition scales teach about machine learning

• The 10 most important lessons 20 years of mathematics taught me

And finally, whenever you’re ready…

If you want to stop wasting time figuring out the math of machine learning on your own, I wrote the perfect book for you—so you can cut through the noise and learn 10X faster.

Check out my bestselling book on Amazon.