The design of a minimum risk classifier based on data usually stems from the fundamental assumption that the operating conditions are the same as the assumed when learning the classifier: the misclassification costs assumed during training must be in agreement with real costs, and the same statistical process must have generated both training and test data. Unfortunately, in real world applications, these assumptions may not hold. This work deals with the problem of training a classifier when prior probabilities cannot be reliably induced from training data.

To protect against this uncertainty, we propose a minimax regret (minimax deviation) approach that seeks for minimizing the maximum difference with respect to the optimal risk classifier. A neural-based minimax regret classifier for general multi-class decision problems is presented. Experimental results show its robustness and the advantages in relation to other approaches.

This is a joint work with Jesus Cid-Sueiro and Alicia Guerrero-Curieses