Types of Machine Learning

⏰ 20 min 📚 Beginner 🧠 Core Concept

Section 1

The Question That Defines Every ML Problem

Before you write a single line of code, every machine learning problem starts with one question: what kind of information do you actually have, and what are you trying to do with it? The answer sorts your problem into one of a small number of categories, and that category determines which entire family of techniques applies. Get this categorization wrong, and you'll waste time forcing the wrong tool at a problem — this is one of the most common mistakes beginners make.

Section 2

Supervised Learning

You have labeled examples — inputs paired with known correct answers — and you want the model to learn the relationship well enough to predict answers for new, unseen inputs.

The pattern Input (X) + Known Answer (Y) \to Model learns X \to Y \to Predict Y for new X

Two flavors of supervised learning, depending on what kind of answer you're predicting:

🔢

Classification

The answer is a category. "Is this email spam or not spam?" "Is this tumor benign or malignant?" You'll build your first classifier in Module 7 of this tier.

📊

Regression

The answer is a number. "What will this house sell for?" "How many units will we sell next month?" Also covered hands-on in Module 7.

💡

Supervised learning is, by far, the most common type used in real business applications — almost everything in Tier 2 falls into this category. The catch is that it requires labeled data, which is often the hardest and most expensive part of a real ML project, far more than the modeling itself.

Section 3

Unsupervised Learning

You have data but no labeled answers — nobody has told the model what the "correct" output should be. Instead, the model finds structure or patterns hidden in the data on its own.

📊

Clustering

Group similar data points together without being told what the groups should be. Example: automatically segmenting customers into behavior groups for marketing. Covered in Tier 2, Module 4.

📊

Dimensionality Reduction

Compress complex, high-dimensional data down to its essential structure for visualization or further processing. Covered in Tier 2, Module 5.

Section 4

Reinforcement Learning

An agent takes actions in an environment and learns from rewards or penalties over time — closer to how a dog learns tricks through treats than to the example-based learning above. This is the framework behind game-playing AI (like AlphaGo) and is also the technique behind how modern chatbots get fine-tuned to be more helpful (RLHF — reinforcement learning from human feedback, which you'll meet again in Tier 4).

Agent

Takes an action

→

Environment

Responds with new state

→

Reward Signal

Tells agent how well it did

→

Agent Adjusts

Improves future actions

RL is covered in depth in Tier 3, Module 10 — it's a smaller part of this track than supervised learning, but it underlies some of the most impressive AI demonstrations you've likely seen (game-playing agents, robotics).

Section 5

Self-Supervised Learning

This is the newest category to become mainstream, and it's the one that quietly powers the entire LLM revolution. The idea: take unlabeled data (like all the text on the internet) and automatically generate the "labels" from the data itself — for example, hide a word in a sentence and train the model to predict it. No human ever had to label anything.

🔭

This is precisely why large language models could be trained on such enormous amounts of text: self-supervised learning sidesteps the single biggest bottleneck in supervised learning — the cost and effort of human labeling. You'll see this mechanism in detail in Tier 4, Module 2.

Section 6

Putting It All Together

Type	Has Labels?	Goal	Where in This Track
Supervised	Yes	Predict an answer for new inputs	Tier 1-2 (most of the course)
Unsupervised	No	Find hidden structure/patterns	Tier 2, Modules 4-5
Reinforcement	Rewards instead	Learn good actions through trial and feedback	Tier 3, Module 10
Self-Supervised	Auto-generated	Learn from unlabeled data at massive scale	Tier 4, Module 2

🎉 Lesson 1.4 complete! Next: real career paths in this field.