Summary  
This chapter covers the three main machine learning paradigms: supervised learning for regression and classification, unsupervised learning for clustering, anomaly detection, and dimensionality reduction, and reinforcement learning based on rewards and penalties.

General domain of usage  
Autonomous driving

**Supervised learning** is a machine learning technique in which the model is trained on a **labeled training set**.

Definition

The most popular supervised learning tasks are:
- **Regression** (for example, predicting the price of a house): you will need a training set labeled with other house prices for that;

- **Classification** (for example, classifying email as spam/ham): you will need a training set labeled as spam/ham for that.

**Unsupervised learning** is a machine learning technique in which the model is trained on an **unlabeled training set**.

The main unsupervised learning tasks are **clustering**, **anomaly detection**, and **dimensionality reduction**.

### Clusterization

Groups similar data points into clusters **without labels** — for example, grouping emails without knowing whether they are spam or not.

### Anomaly Detection

Finds data points that **deviate** from normal patterns, such as unusual credit card transactions, without needing fraud labels.

### Dimensionality Reduction

Reduces the number of features while preserving important information — also label-free.

## Reinforcement Learning

Reinforcement learning differs significantly from the previous two types. It is a technique used to train self-driving vehicles, robots, AI in gaming, and more.

**Reinforcement learning** is a machine learning technique in which the **agent** (e.g., vacuum cleaner robot) learns by making decisions and getting a **reward** if the decision is correct and a **penalty** if the decision is wrong.

Training a dog to fetch works similarly to reinforcement learning:
good actions earn a **reward**, wrong actions earn a **penalty**, and successfully bringing the ball earns a **larger reward**, reinforcing the desired behavior.

To train the ML model for a supervised learning task, you need a training set to contain target (be labeled). Is it correct?

To train the ML model for a unsupervised learning task, containing a target (being labeled) for a training set is not required. Is it correct?

Master the fundamentals of Machine Learning and the Scikit-learn library. Explore the complete ML workflow, from handling missing values and encoding categorical data to scaling features. Build efficient, leak-proof data preprocessing pipelines using ColumnTransformer. Transform raw datasets into model-ready structures and implement robust predictive pipelines.

Types of Machine Learning

Supervised Learning

Unsupervised Learning

Clusterization

Anomaly Detection

Dimensionality Reduction

Reinforcement Learning

1. To train the ML model for a supervised learning task, you need a training set to contain target (be labeled). Is it correct?

2. To train the ML model for a unsupervised learning task, containing a target (being labeled) for a training set is not required. Is it correct?