KuLSIF: Kernel-based Unconstrained Least-Squares Importance Fitting

KuLSIF is a kernel-based learning algorithm to directly estimate the ratio of two density functions without going through density estimation.
R implementation of KuLSIF:

Robust Parameter Fitting using Scoring Rules

This code is supplementary material for
T. Kanamori, H, Fujisawa: "Robust Estimation under Heavy Contamination using Unnormalized Models", in Biometrika.
R implementation:
Go to Kanamori's web site