Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 392 - Big Tensor Data Analysis
Type: Invited
Date/Time: Wednesday, August 5, 2020 : 1:00 PM to 2:50 PM
Sponsor: Section on Statistical Learning and Data Science
Abstract #309286
Title: Co-Clustering Tensors Using Fusion Penalties and CP-Decompositions
Author(s): Eric Chi*
Companies: North Carolina State University
Keywords: Tensor; Candecomp/Parafac; Co-clustering; variable selection; optimization; low-rank

In the tensor co-clustering problem, we seek to organize subarrays along the modes of the tensors into groups with similar signal values. We introduce a new co-clustering estimator that is similar in spirit to a recently introduced convex formulation of the co-clustering problem but employs a CP-decomposition to expedite computation. We demonstrate that in practice our new estimator has comparable performance to existing co-clustering methods but is computationally more scalable.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program