In “Active Learning Literature Survey“, Burr Settles (2010) reviews uncertainty sampling (Lewis and Gale, 1994), margin sampling (Scheffer et al., 2001), entropy sampling, optimal experimental design, query-by-committee (Seung et al., 1992), query-by-boosting, query-by-bagging, expected model change, expected error reduction, expected information gain, variance reduction, and density weighted methods.  He then comments on theoretical and empirical performance of these methods, practical considerations, and related areas of machine learning including: semi-supervised learning, reinforcement learning, and compression.

Related Posts via Categories

• AlphaGo is changing how the Game is Played
• Counterfactual Regret Minimization
• Link: In Two Moves, AlphaGo and Lee Sedol Redefined the Future
• A Short Informal Description of Partially Observable Markov Decision Processes
• University of Alberta Solves Poker
• A review of “Playing Atari with Deep Reinforcement Learning”
• An AI for 2048 — Part 1
• “Generalized Thompson sampling for sequential decision-making and causal inference”
• Reinforcement Learning for Linear Control Systems
• The 20 most striking papers, workshops, and presentations from NIPS 2012