1: \begin{abstract}   

2:         We present an alternating least squares type numerical optimization scheme to estimate conditionally-independent mixture models in $\mathbb{R}^n$, with minimal additional distributional assumptions.  

3:         Following the method of moments, we tackle a coupled system of low-rank tensor decomposition problems.  

4:         The steep costs associated with high-dimensional tensors are avoided, through the development of specialized tensor-free operations.

5:         Numerical experiments illustrate the performance of the algorithm and its applicability to various  models and applications.

6:         In many cases the results exhibit improved reliability over the expectation-maximization algorithm, with similar time and storage costs.

7:         We also provide some supporting theory, establishing identifiability and local linear convergence.

8:     \end{abstract}