Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Common and Individual Feature Extraction using Tensor Decompositions: A Remedy for the Curse of Dimensionality?

Abstract: 

A novel method for common and individual feature analysis from exceedingly large-scale data is proposed, in order to ensure the tractability of both the computation and storage and thus mitigate the curse of dimensionality, a major bottleneck in modern data science. This is achieved by making use of the inherent redundancy in so-called multi-block data structures, which represent multiple observations of the same phenomenon taken at different times, angles or recording conditions. Upon providing an intrinsic link between the properties of the outer vector product and extracted features in tensor decompositions (TDs), the proposed common and individual information extraction from multi-block data is performed through constraints which impose physical meaning on otherwise unconstrained factorisation approaches. This is shown to dramatically reduce the dimensionality of search spaces in subsequent classification procedures and to yield greatly enhanced accuracy. Simulations on a multi-class classification task of large-scale extraction of individual features from a collection of partially related real-world images demonstrate the advantages of the ``blessing of dimensionality'' associated with TDs.

up
0 users have voted:

Paper Details

Authors:
Ilia Kisil, Giuseppe G. Calvi, Andrzej Cichocki, Danilo P. Mandic
Submitted On:
13 April 2018 - 2:12pm
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
ILIA KISIL
Paper Code:
4478
Document Year:
2018
Cite

Document Files

KISIL_ICASSP_2018.pdf

(43 downloads)

Subscribe

[1] Ilia Kisil, Giuseppe G. Calvi, Andrzej Cichocki, Danilo P. Mandic, "Common and Individual Feature Extraction using Tensor Decompositions: A Remedy for the Curse of Dimensionality?", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2739. Accessed: May. 25, 2018.
@article{2739-18,
url = {http://sigport.org/2739},
author = {Ilia Kisil; Giuseppe G. Calvi; Andrzej Cichocki; Danilo P. Mandic },
publisher = {IEEE SigPort},
title = {Common and Individual Feature Extraction using Tensor Decompositions: A Remedy for the Curse of Dimensionality?},
year = {2018} }
TY - EJOUR
T1 - Common and Individual Feature Extraction using Tensor Decompositions: A Remedy for the Curse of Dimensionality?
AU - Ilia Kisil; Giuseppe G. Calvi; Andrzej Cichocki; Danilo P. Mandic
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2739
ER -
Ilia Kisil, Giuseppe G. Calvi, Andrzej Cichocki, Danilo P. Mandic. (2018). Common and Individual Feature Extraction using Tensor Decompositions: A Remedy for the Curse of Dimensionality?. IEEE SigPort. http://sigport.org/2739
Ilia Kisil, Giuseppe G. Calvi, Andrzej Cichocki, Danilo P. Mandic, 2018. Common and Individual Feature Extraction using Tensor Decompositions: A Remedy for the Curse of Dimensionality?. Available at: http://sigport.org/2739.
Ilia Kisil, Giuseppe G. Calvi, Andrzej Cichocki, Danilo P. Mandic. (2018). "Common and Individual Feature Extraction using Tensor Decompositions: A Remedy for the Curse of Dimensionality?." Web.
1. Ilia Kisil, Giuseppe G. Calvi, Andrzej Cichocki, Danilo P. Mandic. Common and Individual Feature Extraction using Tensor Decompositions: A Remedy for the Curse of Dimensionality? [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2739