CaltechAUTHORS
  A Caltech Library Service

Multi-resolution Tensor Learning for Large-Scale Spatial Data

Zheng, Stephan and Yu, Rose and Yue, Yisong (2018) Multi-resolution Tensor Learning for Large-Scale Spatial Data. . (Submitted) http://resolver.caltech.edu/CaltechAUTHORS:20181023-101356776

[img] PDF - Submitted Version
See Usage Policy.

2435Kb

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20181023-101356776

Abstract

High-dimensional tensor models are notoriously computationally expensive to train. We present a meta-learning algorithm, MMT, that can significantly speed up the process for spatial tensor models. MMT leverages the property that spatial data can be viewed at multiple resolutions, which are related by coarsening and finegraining from one resolution to another. Using this property, MMT learns a tensor model by starting from a coarse resolution and iteratively increasing the model complexity. In order to not "over-train" on coarse resolution models, we investigate an information-theoretic fine-graining criterion to decide when to transition into higher-resolution models. We provide both theoretical and empirical evidence for the advantages of this approach. When applied to two real-world large-scale spatial datasets for basketball player and animal behavior modeling, our approach demonstrate 3 key benefits: 1) it efficiently captures higher-order interactions (i.e., tensor latent factors), 2) it is orders of magnitude faster than fixed resolution learning and scales to very fine-grained spatial resolutions, and 3) it reliably yields accurate and interpretable models.


Item Type:Report or Paper (Discussion Paper)
Related URLs:
URLURL TypeDescription
https://arxiv.org/abs/1802.06825arXivDiscussion Paper
Additional Information:This result is supported in part by NSF #1564330, NSF #1637598, and gifts from Bloomberg and Northrop Grumman.
Funders:
Funding AgencyGrant Number
NSFCCF-156433
NSFCCF-1637598
BloombergUNSPECIFIED
Northrop Grumman CorporationUNSPECIFIED
Record Number:CaltechAUTHORS:20181023-101356776
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20181023-101356776
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:90356
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:23 Oct 2018 19:52
Last Modified:23 Oct 2018 19:52

Repository Staff Only: item control page