Home▸AI & Machine Learning▸Query-by-Committee: Vote-Entropy Active Learning

Query-by-Committee: Vote-Entropy Active Learning

Interactive 3D simulator of Query-by-Committee active learning: a committee of logistic-regression classifiers votes on unlabeled points, vote entropy picks the next label to request, and the decision boundaries converge as you watch.

AI & Machine Learning3DAdvanced60 FPS📱 Mobile-adapted⇄ 2D version

Active learning tries to reach high accuracy from as few labeled examples as possible by choosing, at every step, which unlabeled point is worth asking a human to label. This simulator implements Query-by-Committee (QBC): a committee of logistic-regression classifiers is trained on bootstrap resamples of the labeled set, every unlabeled pool point gets a vote from each member, and the point with the highest vote entropy — the one the committee disagrees about most — is queried next. Watch the decision boundaries converge and the committee accuracy climb as labels are added, and compare against a random-query baseline to see how much the disagreement-driven strategy saves.

⚙ Under the hood

Watch a committee of logistic-regression classifiers vote on unlabeled points and pick the next label to request by vote entropy — the classic Query-by-Committee active-learning strategy, with a random-query baseline to compare against.

active learningquery by committeemachine learninglogistic regressionvote entropydata science

3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install