Hierarchical Quadratic Random Forest Classifier
This is an incremental method for improving classification in domains with complex data structures, such as medical imaging or remote sensing.
The paper tackled the problem of classifying multiresolution samples from multichannel data by proposing a hierarchical quadratic random forest classifier, which achieved quadratic decision boundaries through penalized multivariate linear discriminants and group Lasso regularization.
In this paper, we proposed a hierarchical quadratic random forest classifier for classifying multiresolution samples extracted from multichannel data. This forest incorporated a penalized multivariate linear discriminant in each of its decision nodes and processed squared features to realize quadratic decision boundaries in the original feature space. The penalized discriminant was based on a multiclass sparse discriminant analysis and the penalization was based on a group Lasso regularizer which was an intermediate between the Lasso and the ridge regularizer. The classification probabilities estimated by this forest and the features learned by its decision nodes could be used standalone or foster graph-based classifiers.