Living Reference Work Entry

Encyclopedia of Algorithms

pp 1-6

Date: Latest Version

Active Learning - Modern Learning Theory

  • Maria-Florina BalcanAffiliated withDepartment of Machine Learning, Carnegie Mellon University Email author 
  • , Ruth UrnerAffiliated withDepartment of Machine Learning, Carnegie Mellon University


Active learning Learning theory Sample complexity Computational complexity

Years and Authors of Summarized Original Work

2006; Balcan, Beygelzimer, Langford

2007; Balcan, Broder, Zhang

2007; Hanneke

2013; Urner, Wulff, Ben-David

2014; Awashti, Balcan, Long

Problem Definition

Most classic machine learning methods depend on the assumption that humans can annotate all the data available for training. However, many modern machine learning applications (including image and video classification, protein sequence classification, and speech processing) have massive amounts of unannotated or unlabeled data. As a consequence, there has been tremendous interest both in machine learning and its application areas in designing algorithms that most efficiently utilize the available data while minimizing the need for human intervention. An extensively used and studied technique is active learning, where the algorithm is presented with a large pool of unlabeled examples (such as all images available on the web) and can interactively ask for the labels of examples of its own choosing ...

This is an excerpt from the content