Feature set partitioning generalizes the task of feature selection by partitioning the feature set into subsets of features that are collectively useful, rather than by finding a ...
This work studies the problem of design of nonlinear observers in the presence of exogenous disturbances. In particular, the present work proposes a systematic design method for no...
The traditional power spectral subtraction algorithm is computationally simple to implement but suffers from musical noise distortion. In addition, the subtractive rules are based...
The concepts of similarity and distance are crucial in data mining. We consider the problem of defining the distance between two data sets by comparing summary statistics compute...
In any real-life identification problem, only a finite number of data points is available. On the other hand, almost all results in stochastic identification pertain to asymptotic...