Smoothed analysis of tensor decompositions

Aditya Bhaskara, Moses Charikar, Ankur Moitra, Aravindan Vijayaraghavan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

65 Scopus citations


Low rank decomposition of tensors is a powerful tool for learning generative models. The uniqueness results that hold for tensors give them a significant advantage over matrices. However, tensors pose serious algorithmic challenges; in particular, much of the matrix algebra toolkit fails to generalize to tensors. Efficient decomposition in the overcomplete case (where rank exceeds dimension) is particularly challenging. We introduce a smoothed analysis model for studying these questions and develop an efficient algorithm for tensor decomposition in the highly overcomplete case (rank polynomial in the dimension). In this setting, we show that our algorithm is robust to inverse polynomial error-a crucialproperty for applications in learning since we are only allowed a polynomial number of samples. While algorithms are known for exact tensor decomposition in some overcomplete settings, our main contribution is in analyzing theirstability in the framework of smoothed analysis. Our main technical contribution is to show that tensor products of perturbed vectors are linearly independent in a robust sense (i.e. the associated matrix has singular values that are at least an inverse polynomial). This key result paves the way for applying tensor methods to learning problems in the smoothed setting. In particular, we use it to obtain results for learning multi-view models and mixtures of axis-aligned Gaussians where there are many more "components" than dimensions. The assumption here is that the model is not adversarially chosen, which we formalize by thinking of the model parameters as being perturbed. We believe this an appealing way to analyze realistic instances of learning problems, since this framework allows us to overcome many of the usual limitations of using tensor methods.

Original languageEnglish (US)
Title of host publicationSTOC 2014 - Proceedings of the 2014 ACM Symposium on Theory of Computing
PublisherAssociation for Computing Machinery
Number of pages10
ISBN (Print)9781450327107
StatePublished - 2014
Event4th Annual ACM Symposium on Theory of Computing, STOC 2014 - New York, NY, United States
Duration: May 31 2014Jun 3 2014

Publication series

NameProceedings of the Annual ACM Symposium on Theory of Computing
ISSN (Print)0737-8017


Other4th Annual ACM Symposium on Theory of Computing, STOC 2014
Country/TerritoryUnited States
CityNew York, NY

ASJC Scopus subject areas

  • Software


Dive into the research topics of 'Smoothed analysis of tensor decompositions'. Together they form a unique fingerprint.

Cite this