Adaptive Local Linear Regression With Application to Printer Color Management

Abstract

Local learning methods, such as local linear regression and nearest neighbor classifiers, base estimates on nearby training samples, neighbors. Usually, the number of neighbors used in estimation is fixed to be a global "optimal" value, chosen by cross validation. This paper proposes adapting the number of neighbors used for estimation to the local geometry of the data, without need for cross validation. The term enclosing neighborhood is introduced to describe a set of neighbors whose convex hull contains the test point when possible. It is proven that enclosing neighborhoods yield bounded estimation variance under some assumptions. Three such enclosing neighborhood definitions are presented: natural neighbors, natural neighbors inclusive, and enclosing k-NN. The effectiveness of these neighborhood definitions with local linear regression is tested for estimating lookup tables for color management. Significant improvements in error metrics are shown, indicating that enclosing neighborhoods may be a promising adaptive neighborhood definition for other local learning tasks as well, depending on the density of training samples.

Rotation invariant texture descriptors based on Gaussian Markov random fields for classification

“…Parameter estimation stage of the TGMRF descriptors suffers from producing estimates that are biased and over-smoothed when the GMRF model do not capture the underlying data generating process (Gupta et al, 2008). This reduce the texture discriminative power of TGMRF features.…”

Virtual Sensing Technology in Process Industries: Trends and Challenges Revealed by Recent Industrial Applications

“…Another approach is to build a local model as a Taylor series of the global model expanded in the vicinity of the query (Su et al, 2012). Gupta et al (2008) proved that the variance of estimates of LLR is bounded by the variance of measurement noise if a query is in the convex hull enclosing its neighborhood.…”

Local regression transfer learning with applications to users’ psychological characteristics prediction

“…Local machine learning has shown a comparative advantage in many machine learning tasks [15–17]. In some situations, the size of local region of target data imposes a significant effect on prediction accuracy of model [17].…”