Skip to main content
learned

learned

6.2
Made by a computer from data rather than designed by a person
  • adjective
  • /ˈlɜrnɪd/
  • Specialized
translation icon : aprendido
  • Engineers fine-tuned the system until the learned parameters generalized well to unseen test examples.

And in some cases it is building a learned capability.

Examples

  • The model's learned embeddings captured subtle relationships between words across multiple languages.

  • Rather than handcrafting features, the team relied on learned representations extracted automatically from raw sensor data.

  • The learned model improved its accuracy significantly.

  • This algorithm produces learned representations from vast datasets.

  • The model's learned features improved the prediction accuracy in the test data.

  • The software generated learned representations based on millions of training samples.

  • After extensive training, the AI developed a learned understanding of user preferences.

Antonyms

101 3.8

Done by hand rather than by a machine

Compounds

1 4.4

A person who knows a lot about a subject from study

learned response
  • Specialized
1 5.7

A reaction a person develops from practice or experience, not one they are born with

6.4

A job like law or medicine that needs a lot of study and high moral standards

learned reaction
  • Specialized
5.3

A way of acting you get from practice or experience

How Learned

trained
1.9
learned
  • Specialized
6.2
untrained
31 3.4

Surface Forms

learned positive

Morphology

learned = learn (transparent) = learn + ed

In technical contexts 'learned' means 'produced/obtained by learning' (learn + -ed), a straightforward participial adjective formation.

Etymology

Learned in machine learning keeps the same root learn meaning 'to get knowledge', but now it often means 'made by a computer program from data', like a child learning from many examples. So a learned model or feature is one the system built by studying data rather than a person designing it.