What is it about?
This work presents a methodology to estimate the value of the Epanechnikov kernel between two feature vectors counting on missing entries.
Featured Image
Why is it important?
The results presented here can be used as tools to adapt methods using the Epanechnikov kernel so they can handle missing/incomplete data.
Perspectives
Assuming the squared distance between two incomplete feature vectors can be modelled after a gamma distribution, this work presents an elegant formulation to estimate the Epanechnikov kernel as a transform of this squared distance.
Diego Mesquita
Aalto-yliopisto
Read the Original
This page is a summary of: Epanechnikov Kernel for Incomplete Data , Electronics Letters, August 2017, the Institution of Engineering and Technology (the IET),
DOI: 10.1049/el.2017.0507.
You can read the full text:
Contributors
The following have contributed to this page
