Tensor sparsification via a bound on the spectral norm of random tensors

Abstract

Given an order-$d$ tensor $\tensor A \in \R^{n \times n \times...\times n}$, we present a simple, element-wise sparsification algorithm that zeroes out all sufficiently small elements of $\tensor A$, keeps all sufficiently large elements of $\tensor A$, and retains some of the remaining elements with probabilities proportional to the square of their magnitudes. We analyze the approximation accuracy of the proposed algorithm using a powerful inequality that we derive. This inequality bounds the spectral norm of a random tensor and is of independent interest. As a result, we obtain novel bounds for the tensor sparsification problem.

33 pages; Information and Inference (A journal of the IMA), 2014

References in corpus (2)

• MACH: Fast Randomized Tensor Decompositions
• Error Bounds for Random Matrix Approximation Schemes

Cited by in corpus (6)

• A Note on Element-wise Matrix Sparsification via a Matrix-valued Bernstein Inequality
• A New Sampling Technique for Tensors
• Sample Complexity Analysis for Learning Overcomplete Latent Variable Models through Tensor Methods
• Quantum Machine Learning Algorithm for Knowledge Graphs
• Spatio-Temporal Tensor Sketching via Adaptive Sampling
• Randomized Interpolative Decomposition of Separated Representations