1: \begin{abstract}

2: 	We present a fast variational Bayesian algorithm for performing non-negative matrix factorisation and tri-factorisation.

3:  	We show that our approach achieves faster convergence per iteration and timestep (wall-clock) than Gibbs sampling and non-probabilistic approaches, and do not require additional samples to estimate the posterior.

4:  	We show that in particular for matrix tri-factorisation convergence is difficult, but our variational Bayesian approach offers a fast solution, allowing the tri-factorisation approach to be used more effectively.

5: \end{abstract}

6: