Evolutive Approaches for Variable Selection Using a Non-parametric Noise Estimator

摘要: The design of a model to approximate function relies significantly on the data used in training stage. problem selecting an adequate set variables should be treated carefully due its importance. If number is high, samples needed becomes too large and interpretability lost. This chapter presents several methodologies perform variable selection local or globalmanner using non-parametric noise estimator determine quality subset variables. Several methods that apply parallel paradigms different architecures are compared from optimization efficiency point view since computationally expensive.