Active Learning

Fact

A way to train models by labeling only the most useful examples.

In Plain Words

Active Learning is like a kid in math class. They skip 2+2 and ask the teacher about the monster problem.

It helps when labels cost money. The model asks people about fewer examples, but learns faster.

Related Concepts

• Supervised Learning
• Data Labeling
• Model uncertainty
• Human-in-the-loop

Supervised Learning
Active Learning picks more useful examples for Supervised Learning.

Data Labeling
Active Learning spends the labeling budget on examples worth asking about.

Model uncertainty
The model asks first about examples it is unsure about.

Human-in-the-loop
Active Learning needs people to label the key examples.