Papers
Topics
Authors
Recent
Search
2000 character limit reached

Guaranteed Simultaneous Asymmetric Tensor Decomposition via Orthogonalized Alternating Least Squares

Published 25 May 2018 in stat.ML and cs.LG | (1805.10348v2)

Abstract: Tensor CANDECOMP/PARAFAC (CP) decomposition is an important tool that solves a wide class of machine learning problems. Existing popular approaches recover components one by one, not necessarily in the order of larger components first. Recently developed simultaneous power method obtains only a high probability recovery of top rr components even when the observed tensor is noiseless. We propose a Slicing Initialized Alternating Subspace Iteration (s-ASI) method that is guaranteed to recover top rr components (ϵ\epsilon-close) simultaneously for (a)symmetric tensors almost surely under the noiseless case (with high probability for a bounded noise) using O(log⁡(log⁡1ϵ))O(\log(\log \frac{1}{\epsilon})) steps of tensor subspace iterations. Our s-ASI method introduces a Slice-Based Initialization that runs O(1/log⁡(λrλr+1))O(1/\log(\frac{\lambda_r}{\lambda_{r+1}})) steps of matrix subspace iterations, where λr\lambda_r denotes the r-th top singular value of the tensor. We are the first to provide a theoretical guarantee on simultaneous orthogonal asymmetric tensor decomposition. Under the noiseless case, we are the first to provide an \emph{almost sure} theoretical guarantee on simultaneous orthogonal tensor decomposition. When tensor is noisy, our algorithm for asymmetric tensor is robust to noise smaller than min⁡{O((λr−λr+1)ϵr),O(δ0λr−λr+1d)}\min\{O(\frac{(\lambda_r - \lambda_{r+1})\epsilon}{\sqrt{r}}), O(\delta_0\frac{\lambda_r -\lambda_{r+1}}{\sqrt{d}})\}, where δ0\delta_0 is a small constant proportional to the probability of bad initializations in the noisy setting.

Authors (3)
Citations (5)

Summary

No one has generated a summary of this paper yet.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Continue Learning

We haven't generated follow-up questions for this paper yet.