Neural network

Your learning move

OrientClarify a term, source, person, or event.

Why this page exists

Reference pages support learning without forcing you to leave your current question or path.

What you will leave with

A neural network is a parameterized computing structure that passes information through connected units and transformations to learn an input–output relationship.

Before you begin

You can use these pages independently or return to the learning object that sent you here.

Do this now

Find the one fact or relationship you need, then return to your original learning goal.

Do now on this pageClarify this conceptThe plan states the outcome, time, prerequisite, immediate action, and prepared continuation.Another valid routeInspect SourcesSee provenance, review, and evidence boundaries.

Account

Connected glossary

Neural network

A neural network is a parameterized computing structure that passes information through connected units and transformations to learn an input–output relationship.

Plan for this page

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

You will leave with: Training adjusts many numerical parameters so that the network better serves an objective on examples. Modern networks can have different architectures for images, language, audio, control, or combinations of modalities.
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 stepSee how models are trained and used

Neural networkA neural network is a parameterized computing structure that passes information through connected units and transformations to learn an input–output relationship.

Deep learning
Parameters
Feature

Plain explanation

Training adjusts many numerical parameters so that the network better serves an objective on examples. Modern networks can have different architectures for images, language, audio, control, or combinations of modalities.

Example

A network trained on handwritten digits receives pixel values and produces a probability for each digit from zero to nine.

Why it matters

Neural networks support much of modern AI, but the architecture name does not establish reliability or understanding. Data, objective, training, evaluation, and deployment context remain decisive.

Common misunderstanding

Misconception: an artificial neural network is a digital copy of the human brain. Correction: it is a mathematical computing design inspired only loosely by some ideas about connected neurons.

Related concepts

Follow relationships, not only the alphabet

Deep learningDeep learning is machine learning built with neural networks containing multiple layers of learned transformations.ParametersParameters are internal numerical values that a model’s training procedure adjusts to shape its behavior.FeatureA feature is an input property or representation used by a model when producing an output.

See how models are trained and used

Sources and scope

What supports this object

Machine Learning GlossaryGoogle for Developers

Defines neural-network components in a technical learning context; biological analogies are limited and should not imply equivalence to brains.

Open source

Loading page content...

…