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:
- kernelDRest.r computes kernel-based uLSIF (KuLSIF).
required library: "kernlab".
"snow" is available for parallel computing.
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