Optimally Combining Classifiers Using Unlabeled Data
Proceedings of The 28th Conference on Learning Theory, PMLR 40:211-225, 2015.
We develop a worst-case analysis of aggregation of classifier ensembles for binary classification. The task of predicting to minimize error is formulated as a game played over a given set of unlabeled data (a transductive setting), where prior label information is encoded as constraints on the game. The minimax solution of this game identifies cases where a weighted combination of the classifiers can perform significantly better than any single classifier.