Browse Definitions :
Definition

active learning (active learner algorithm)

Contributor(s): Ivy Wigmore

Active learning, in an AI context, is the capacity of a machine learning (ML) algorithm to query a human source for additional information. Essentially, active learning allows a minimally-trained AI program to identify what data it needs to become better. The algorithm identifies which subset of data it expects to yield the best results for a particular category and requests that someone label the data in that subset. 

Active learning algorithms require minimal training data, which makes them especially helpful when there’s not a lot of labeled data available. This makes this type of algorithm useful for information retrieval and text analysis -- as well as image and speech recognition.

Active learning vs. supervised learning vs. unsupervised learning

Active learning algorithms are a simple form of semi-supervised, curious AI. This type of algorithm combines aspects of both supervised learning and unsupervised learning.

Supervised ML, which uses historic data to make forecasts about new data, requires a human to create input and desired-output data for training. Because this approach requires a lot of human overhead, it can be expensive. In contrast, AI systems that use unsupervised learning require very little human overhead because the algorithms simply look for patterns in unlabeled datasets. While this type of ML can be cost-effective because it does not require as much human input, it can also be difficult to quantify the results as being meaningful. 

Active learning can use both structured and unstructured data in a cost-efficient manner by prioritizing which data the model is most confused about and requesting labels for just that data. The model will use a relatively small amount of labeled data for training and request more labels later on if needed. This iterative approach to machine learning not only helps the machine learning model learn faster, it also keeps costs down by letting humans skip labeling data that isn't helpful to the model.

Watch this introductory tutorial on active learning to learn more.

This was last updated in December 2019

Continue Reading About active learning (active learner algorithm)

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.

-ADS BY GOOGLE

File Extensions and File Formats

SearchCompliance

  • risk assessment

    Risk assessment is the identification of hazards that could negatively impact an organization's ability to conduct business.

  • PCI DSS (Payment Card Industry Data Security Standard)

    The Payment Card Industry Data Security Standard (PCI DSS) is a widely accepted set of policies and procedures intended to ...

  • risk management

    Risk management is the process of identifying, assessing and controlling threats to an organization's capital and earnings.

SearchSecurity

SearchHealthIT

  • telemedicine (telehealth)

    Telemedicine is the remote delivery of healthcare services, such as health assessments or consultations, over the ...

  • Project Nightingale

    Project Nightingale is a controversial partnership between Google and Ascension, the second largest health system in the United ...

  • medical practice management (MPM) software

    Medical practice management (MPM) software is a collection of computerized services used by healthcare professionals and ...

SearchDisasterRecovery

SearchStorage

  • hot plugging

    Hot plugging is the addition of a component to a running computer system without significant interruption to the operation of the...

  • M.2 SSD

    An M.2 SSD is a solid-state drive (SSD) that conforms to a computer industry specification and is used in internally mounted ...

  • kilobyte (KB or Kbyte)

    A kilobyte (KB or Kbyte) is a unit of measurement for computer memory or data storage used by mathematics and computer science ...

Close