Module 1 · Lesson 1

AI, machine learning, and deep learning are often used as if they mean the same thing. They do not. The easiest way to understand them is to start with the broadest idea and work inward.

Artificial intelligence is the broader field

Artificial intelligence is the field of building computer systems that can perform tasks we associate with intelligent behavior. That can include perception, planning, language, decision support, search, reasoning, or acting in an environment.

Machine learning is one important way to build AI systems, but it is not the only one. Earlier AI systems often relied heavily on rules, search algorithms, symbolic representations, and expert knowledge written directly by people. Those approaches still matter in many systems today.

Artificial Intelligence
├── Rule-based and symbolic approaches
└── Machine Learning
    └── Deep Learning

Machine learning learns patterns from examples

Traditional software usually starts with rules written by a programmer. Give the program an input, apply the rules, and produce an output.

Input + human-written rules → output

Machine learning changes where some of those rules come from. Instead of describing every pattern explicitly, we give an algorithm data and an objective. Training adjusts the model so that useful patterns are captured in its parameters.

Examples + learning algorithm → trained model
New input + trained model → prediction

Consider spam detection. A hand-written system might block a message whenever it contains a particular phrase. A machine-learning system can learn from many examples of spam and legitimate email, discovering combinations of signals that are difficult to express as a short list of rules.

Deep learning is a type of machine learning

Deep learning uses neural networks with many layers of learned computation. These networks have become especially effective for complex data such as language, images, audio, and video.

Modern large language models are deep-learning systems. They are not a separate category outside machine learning. They sit inside this larger hierarchy.

A useful distinction

If you remember only one thing from this lesson, remember this:

  • AI is the broad field.
  • Machine learning is one major approach within AI.
  • Deep learning is a family of machine-learning techniques based on deep neural networks.

This distinction matters because it prevents a common mistake: assuming every intelligent software system must be a neural network. Sometimes a deterministic rule, search algorithm, database query, or traditional statistical model is exactly the right tool.

Demystifying AI series

← Series overview   ·   How Machines Represent Data as Numbers →