Orthogonalized ALS: A Theoretically Principled Tensor Decomposition Algorithm for Practical Use

03/06/2017
by   Vatsal Sharan, et al.
0

The popular Alternating Least Squares (ALS) algorithm for tensor decomposition is efficient and easy to implement, but often converges to poor local optima---particularly when the weights of the factors are non-uniform. We propose a modification of the ALS approach that is as efficient as standard ALS, but provably recovers the true factors with random initialization under standard incoherence assumptions on the factors of the tensor. We demonstrate the significant practical superiority of our approach over traditional ALS for a variety of tasks on synthetic data---including tensor factorization on exact, noisy and over-complete tensors, as well as tensor completion---and for computing word embeddings from a third-order word tri-occurrence tensor.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
05/25/2018

Guaranteed Simultaneous Asymmetric Tensor Decomposition via Orthogonalized Alternating Least Squares

We consider the asymmetric orthogonal tensor decomposition problem, and ...
research
01/09/2020

Coupled Tensor Completion via Low-rank Tensor Ring

The coupled tensor decomposition aims to reveal the latent data structur...
research
09/14/2023

Decomposition of linear tensor transformations

One of the main issues in computing a tensor decomposition is how to cho...
research
04/10/2017

Word Embeddings via Tensor Factorization

Most popular word embedding techniques involve implicit or explicit fact...
research
11/11/2016

Simple and Efficient Parallelization for Probabilistic Temporal Tensor Factorization

Probabilistic Temporal Tensor Factorization (PTTF) is an effective algor...
research
06/04/2020

Tensor Completion Made Practical

Tensor completion is a natural higher-order generalization of matrix com...
research
04/08/2017

Noisy Tensor Completion for Tensors with a Sparse Canonical Polyadic Factor

In this paper we study the problem of noisy tensor completion for tensor...

Please sign up or login with your details

Forgot password? Click here to reset