Skip to yearly menu bar Skip to main content


Poster

Uncertainty quantification for nonconvex tensor completion: Confidence intervals, heteroscedasticity and optimality

Changxiao Cai · H. Vincent Poor · Yuxin Chen

Keywords: [ Information Theory and Estimation ] [ Non-convex Optimization ] [ Sparsity and Compressed Sensing ] [ Matrix/Tensor Methods ] [ Optimization - Non-convex ]


Abstract:

We study the distribution and uncertainty of nonconvex optimization for noisy tensor completion --- the problem of estimating a low-rank tensor given incomplete and corrupted observations of its entries. Focusing on a two-stage nonconvex estimation algorithm, we characterize the distribution of this estimator down to fine scales. This distributional theory in turn allows one to construct valid and short confidence intervals for both the unseen tensor entries and its underlying tensor factors. The proposed inferential procedure enjoys several important features: (1) it is fully adaptive to noise heteroscedasticity, and (2) it is data-driven and adapts automatically to unknown noise distributions. Furthermore, our findings unveil the statistical optimality of nonconvex tensor completion: it attains un-improvable estimation accuracy --- including both the rates and the pre-constants --- under i.i.d. Gaussian noise.

Chat is not available.