# Hierarchical Methods of Moments

arXiv: https://arxiv.org/abs/1810.07468 · HTML: https://mlsanity.com/paper/1810.07468/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Oct 2018.

## Built on by (most cited first, 1 of 1)

- [Estimating Appearance Models for Image Segmentation via Tensor Factorization](https://mlsanity.com/paper/2208.07853/) (2022)

## Abstract (arXiv)

Spectral methods of moments provide a powerful tool for learning the parameters of latent variable models. Despite their theoretical appeal, the applicability of these methods to real data is still limited due to a lack of robustness to model misspecification. In this paper we present a hierarchical approach to methods of moments to circumvent such limitations. Our method is based on replacing the tensor decomposition step used in previous algorithms with approximate joint diagonalization. Experiments on topic modeling show that our method outperforms previous tensor decomposition methods in terms of speed and model quality.