- [[diversity prediction theorem]], [[crowdsourced fact-checking]], [[many-model thinking]] # Idea Condorcet jury theorem: A majority vote classifies correctly with higher probability than any person (model), and as the number of people (models) becomes large, the accuracy of the majority vote approaches 100%. Assumption: Each of an odd number of people (models) classifies an unknown state of the world as either true or false. Each person (model) classifies correctly with probability $p > 0.5$ and the probability that any person (model) classifies correctly is statistically independent of the correctness of any other person (model). The Condorcet jury theorem is derived from a model constructed to explain the advantages of majority rule. In the model, jurors make binary decisions of guilt or innocence. Each juror is correct more often than not. In order to apply the theorem to collections of models instead of jurors, we interpret each juror's decision as a classification by a model. The [[diversity prediction theorem]] quantifies the contributions of model accuracy and model diversity to the accuracy of many models combined. However, like [[social dilemmas]]—individual rationality leads to collective irrationality. Individuals obey the [[axiom of transitivity]], but collective transitivity can sometimes be violated. # References