Deep learning

Deep learning is machine learning built with neural networks containing multiple layers of learned transformations.

Plan for this page

Understand one concept in plain language and connect it to nearby ideas.

You will leave with: The layers can build increasingly useful internal representations from raw or processed inputs, such as pixels, audio, or tokens. “Deep” describes layered architecture, not wisdom, consciousness, or guaranteed quality.
Time: 3–8 min
Before you begin: No prerequisite. Keep one real example in mind.
Do this now: Read the explanation, then compare the example with the common misunderstanding.

Prepared next stepLearn how data and patterns interact

Deep learningDeep learning is machine learning built with neural networks containing multiple layers of learned transformations.

Machine learning
Neural network
Parameters
Training

Plain explanation

The layers can build increasingly useful internal representations from raw or processed inputs, such as pixels, audio, or tokens. “Deep” describes layered architecture, not wisdom, consciousness, or guaranteed quality.

Example

A vision system may transform pixels through many layers before producing probabilities for objects such as bicycle, dog, or road sign.

Why it matters

Deep learning powers many language, vision, speech, and generative systems. Its capability can be strong while its errors, data dependence, computation, and limited interpretability still require evaluation.

Common misunderstanding

Misconception: deep learning means the model thinks deeply. Correction: depth refers mainly to the number and organization of computational layers.

Related concepts

Follow relationships, not only the alphabet

Machine learningMachine learning is an approach in which a model’s behavior is shaped by finding patterns in data or experience rather than only by fixed task rules.Neural networkA neural network is a parameterized computing structure that passes information through connected units and transformations to learn an input–output relationship.ParametersParameters are internal numerical values that a model’s training procedure adjusts to shape its behavior.TrainingTraining is the process of adjusting a model to improve performance on an objective using data, feedback, or interaction.

Learn how data and patterns interact

Sources and scope

What supports this object

Machine Learning GlossaryGoogle for Developers

Provides technical definitions for deep learning and neural networks; architectures and terminology continue to evolve.

Open source

Loading page content...

…