Orthogonalized ALS: A Theoretically Principled Tensor Decomposition Algorithm for Practical Use
[edit]
Proceedings of the 34th International Conference on Machine Learning, PMLR 70:30953104, 2017.
Abstract
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 nonuniform. 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 overcomplete tensors, as well as tensor completion—and for computing word embeddings from a thirdorder word trioccurrence tensor.
Related Material


