Non-Convex Tensor Recovery from Local Measurements

Abstract: Motivated by the settings where sensing the entire tensor is infeasible, this paper proposes a novel tensor compressed sensing model, where measurements are only obtained from sensing each lateral slice via mutually independent matrices. Leveraging the low tubal rank structure, we reparameterize the unknown tensor ${\boldsymbol {\mathcal X}}<sup>\star$ using two compact tensor factors and formulate the recovery problem as a nonconvex minimization problem. To solve the problem, we first propose an alternating minimization algorithm, termed \textsf{Alt-PGD-Min}, that iteratively optimizes the two factors using a projected gradient descent and an exact minimization step, respectively. Despite nonconvexity, we prove that \textsf{Alt-PGD-Min} achieves $\epsilon$-accuracy recovery with $$\mathcal O\left( \kappa² \log \frac{1}{\epsilon}\right)$$ iteration complexity and $$\mathcal O\left( \kappa^6rn_3\log n_3 \left( \kappa^2r\left(n_1 + n_2 \right) + n_1 \log \frac{1}{\epsilon}\right) \right)$$ sample complexity, where $\kappa$ denotes tensor condition number of $\boldsymbol{\mathcal X}<sup>\star$. To further accelerate the convergence, especially when the tensor is ill-conditioned with large $\kappa$, we prove \textsf{Alt-ScalePGD-Min} that preconditions the gradient update using an approximate Hessian that can be computed efficiently. We show that \textsf{Alt-ScalePGD-Min} achieves $\kappa$ independent iteration complexity $\mathcal O(\log \frac{1}{\epsilon})$ and improves the sample complexity to $$\mathcal O\left( \kappa⁴ rn_3 \log n_3 \left( \kappa^4r(n_1+n_2) + n_1 \log \frac{1}{\epsilon}\right) \right)$$. Experiments validate the effectiveness of the proposed methods.