Rethinking LDA: Moment Matching for Discrete ICA
Abstract
We consider moment matching techniques for estimation in Latent Dirichlet Allocation (LDA). By drawing explicit links between LDA and discrete versions of independent component analysis (ICA), we first derive a new set of cumulant-based tensors, with an improved sample complexity. Moreover, we reuse standard ICA techniques such as joint diagonalization of tensors to improve over existing methods based on the tensor power method. In an extensive set of experiments on both synthetic and real datasets, we show that our new combination of tensors and orthogonal joint diagonalization techniques outperforms existing moment matching methods.
Cite
Text
Podosinnikova et al. "Rethinking LDA: Moment Matching for Discrete ICA." Neural Information Processing Systems, 2015.Markdown
[Podosinnikova et al. "Rethinking LDA: Moment Matching for Discrete ICA." Neural Information Processing Systems, 2015.](https://mlanthology.org/neurips/2015/podosinnikova2015neurips-rethinking/)BibTeX
@inproceedings{podosinnikova2015neurips-rethinking,
title = {{Rethinking LDA: Moment Matching for Discrete ICA}},
author = {Podosinnikova, Anastasia and Bach, Francis and Lacoste-Julien, Simon},
booktitle = {Neural Information Processing Systems},
year = {2015},
pages = {514-522},
url = {https://mlanthology.org/neurips/2015/podosinnikova2015neurips-rethinking/}
}