Classic classifiers
Noisy signal processing methods

In the classification problem, it is assumed that m + 1 hypotheses, ∀ i ∊ {O, ..., m }, H_i, can govern the observation; these hypotheses are generated by a partition, into m + 1 subsets, ∀ i ∊ {O, ...., m }, Θ_i, of the space Θ of parameters θ: the values of θ belonging to the same Θ_i distinguish the different probability laws P_{X I θ} of X that are governed by the same hypothesis H_i.

To classify is to decide, by choosing a decision, ∀ j ∊ {O, ..., m }, δ_j, to which hypothesis,...