Least-Squares Density Ridge Finder (LSDRF) estimates the ridge of the probability density function underlying data. To estimate the ridge, the key approach in LSDRF is to directly estimate the ratios of density-derivatives to its density without going through density estimation. Thanks to the direct approach, LSDRF accurately estimates the density ridges on a variety of data. 

MATLAB implementation of LSDRF: LSDRF.zip

• "demo.m" is a demo file for density ridge estimation.

References

• Hiroaki Sasaki, Takafumi Kanamori and Masashi Sugiyama, "Estimating Density Ridges by Direct Estimation of Density-Derivative-Ratios", the 20th International Conference on Artificial Intelligence and Statistics (AISTATS), Proceedings of Machine Learning Research, vol.54, pp.204–212, 2017. 

• Hiroaki Sasaki, Takafumi Kanamori, Aapo Hyvärinen, Gang Niu and Masashi Sugiyama, "Mode-Seeking Clustering and Density Ridge Estimation via Direct Estimation of Density-Derivative-Ratios", arXiv:1707.01711.