Tensor Density Estimator by Convolution-Deconvolution
Numerical Analysis
2025-10-29 v4 Numerical Analysis
Abstract
We propose a linear algebraic framework for performing density estimation. It consists of three simple steps: convolving the empirical distribution with certain smoothing kernels to remove the exponentially large variance; compressing the empirical distribution after convolution as a tensor train, with efficient tensor decomposition algorithms; and finally, applying a deconvolution step to recover the estimated density from such tensor-train representation. Numerical results demonstrate the high accuracy and efficiency of the proposed methods.
Cite
@article{arxiv.2412.18964,
title = {Tensor Density Estimator by Convolution-Deconvolution},
author = {Yifan Peng and Siyao Yang and Yuehaw Khoo and Daren Wang},
journal= {arXiv preprint arXiv:2412.18964},
year = {2025}
}