AIExplainer
Machine Learning Intermediate 2 min read

What is active learning?

A machine learning approach where the model actively selects the most informative data to learn from

Active learning is a type of machine learning where the model is not just passively given a dataset to learn from, but instead it actively selects the most useful data to learn from, often by asking a human for labels or feedback

Imagine you're trying to learn a new language and you have a teacher who can correct your mistakes. Active learning is like having a student who can ask the teacher to correct the most confusing sentences first, rather than just listening to a random selection of sentences

For example, a company might use active learning to build a model that can classify customer support emails as either positive or negative, by actively selecting the most uncertain emails for a human to label

Active learning is often used in situations where labeling data is expensive or time-consuming, such as in medical imaging or text classification

One common misconception is that active learning is only useful when there is a limited amount of labeled data, but it can also be useful when there is a large amount of unlabeled data

Active learning has its roots in the 1980s, but it has gained popularity in recent years with the development of more advanced machine learning algorithms

query-by-committee selective sampling uncertainty sampling

Three products for different needs — explore what’s relevant to you.