Quick answer · 4 min
Machine learning builds behavior from examples or feedback
What is machine learning?Plan for this page
Get a clear answer, then check that you can use it.
- You will leave with
- Machine learning fits patterns from data; data, objective, evaluation, and context shape what it can do.
- Before you begin
- No prerequisite. Start with your question.
- Do this now
- Read the direct answer before the example and active check.
Prepared next stepLearn data, patterns, and learningMachine learning is a branch of AI in which a model’s behavior is fitted from data, examples, or feedback instead of specifying every rule by hand.
Training adjusts parameters toward an objective; evaluation tests behavior on relevant cases; inference applies the trained model to new inputs.
Example
Spam filtering
A model can learn patterns from messages labeled spam or not spam, then estimate the category of a new message.Important limit
Learning is not human understanding
A model can reproduce useful patterns while inheriting selection errors, labels, objectives, and limits of its training and evaluation.Interactive model
Separate before deciding
Read the complete text alternative
- TrainingThe system adjusts behavior from examples or feedback.
- InferenceThe resulting model produces an output for a new input.
Useful distinctions
Separate the questions before judging
TrainingThe system adjusts behavior from examples or feedback.
InferenceThe resulting model produces an output for a new input.
Active check
What makes a spam classifier a machine-learning system?
Sources and scope
What supports this object
EstablishedMachine Learning Crash CourseGoogle for DevelopersTechnical provider course; terminology and examples do not exhaust the field.
Open source
inAi