Optimal Oracle Inequality for Aggregation of Classifiers Under Low Noise Condition

@inproceedings{Lecu2006OptimalOI,
  title={Optimal Oracle Inequality for Aggregation of Classifiers Under Low Noise Condition},
  author={Guillaume Lecu{\'e}},
  booktitle={COLT},
  year={2006}
}
We consider the problem of optimality, in a minimax sense, and adaptivity to the margin and to regularity in binary classification. We prove an oracle inequality, under the margin assumption (low noise condition), satisfied by an aggregation procedure which uses exponential weights. This oracle inequality has an optimal residual: (log M/n) where κ is the margin parameter, M the number of classifiers to aggregate and n the number of observations. We use this inequality first to construct minimax… CONTINUE READING