Learning with Subset Stacking
Abstract
We propose a new regression algorithm that learns from a set of inputoutput pairs. Our algorithm is designed for populations where the relation between the input variables and the output variable exhibits a heterogeneous behavior across the predictor space. The algorithm starts with generating subsets that are concentrated around random points in the input space. This is followed by training a local predictor for each subset. Those predictors are then combined in a novel way to yield an overall predictor. We call this algorithm ``LEarning with Subset Stacking'' or LESS, due to its resemblance to the method of stacking regressors. We compare the testing performance of LESS with stateoftheart methods on several datasets. Our comparison shows that LESS is a competitive supervised learning method. Moreover, we observe that LESS is also efficient in terms of computation time and it allows a straightforward parallel implementation.
 Publication:

arXiv eprints
 Pub Date:
 December 2021
 DOI:
 10.48550/arXiv.2112.06251
 arXiv:
 arXiv:2112.06251
 Bibcode:
 2021arXiv211206251I
 Keywords:

 Computer Science  Machine Learning;
 Statistics  Machine Learning
 EPrint:
 20 pages, 9 figures, 2 tables