When Does Non-Orthogonal Tensor Decomposition Have No Spurious Local Minima?

Nov 22, 2019
Maziar Sanjabi, Sina Baharlouei, Meisam Razaviyayn, Jason D. Lee



We study the optimization problem for decomposing $d$ dimensional fourth-order Tensors with $k$ non-orthogonal components. We derive \textit{deterministic} conditions under which such a problem does not have spurious local minima. In particular, we show that if $\kappa = \frac{\lambda_{max}}{\lambda_{min}} < \frac{5}{4}$, and incoherence coefficient is of the order $O(\frac{1}{\sqrt{d}})$, then all the local minima are globally optimal. Using standard techniques, these conditions could be easily transformed into conditions that would hold with high probability in high dimensions when the components are generated randomly. Finally, we prove that the tensor power method with deflation and restarts could efficiently extract all the components within a tolerance level $O(\kappa \sqrt{k\tau^3})$ that seems to be the noise floor of non-orthogonal tensor decomposition.



Share this with someone who'll enjoy it:

   Access Paper Source



Share this with someone who'll enjoy it: