Active learning for regression by inverse distance weighting

This paper proposes an active learning (AL) algorithm to solve regression problems based on inverse-distance weighting functions for selecting the feature vectors query. The has following features: (i) supports both pool-based and population-based sampling; (ii) is not tailored a particular class of predictors; (iii) can handle known unknown constraints queryable vectors; (iv) run either sequentially, or in batch mode, depending how often predictor retrained. potentials method are shown numerical tests illustrative synthetic real-world datasets. An implementation algorithm, that we call IDEAL (Inverse-Distance Exploration Active Learning), available at http://cse.lab.imtlucca.it/bemporad/ideal.