Non-negative matrix factorization algorithms greatly improve topic model fits. (arXiv:2105.13440v2 [stat.ML] UPDATED)

Web: http://arxiv.org/abs/2105.13440

We report on the potential for using algorithms for non-negative matrix
factorization (NMF) to improve parameter estimation in topic models. While
several papers have studied connections between NMF and topic models, none have
suggested leveraging these connections to develop new algorithms for fitting
topic models. NMF avoids the "sum-to-one" constraints on the topic model
parameters, resulting in an optimization problem with simpler structure and
more efficient computations. Building on recent advances in optimization
algorithms for NMF, we show that first …

algorithms arxiv factorization ml model negative