TY - GEN
T1 - Riemannian joint dimensionality reduction and dictionary learning on symmetric positive definite manifolds
AU - Kasai, Hiroyuki
AU - Mishra, Bamdev
N1 - Funding Information:
ACKNOWLEDGEMENTS H. Kasai was partially supported by JSPS KAKENHI Grant Numbers JP16K00031 and JP17H01732.
Publisher Copyright:
© EURASIP 2018.
PY - 2018/11/29
Y1 - 2018/11/29
N2 - Dictionary leaning (DL) and dimensionality reduction (DR) are powerful tools to analyze high-dimensional noisy signals. This paper presents a proposal of a novel Riemannian joint dimensionality reduction and dictionary learning (R-JDRDL) on symmetric positive definite (SPD) manifolds for classification tasks. The joint learning considers the interaction between dimensionality reduction and dictionary learning procedures by connecting them into a unified framework. We exploit a Riemannian optimization framework for solving DL and DR problems jointly. Finally, we demonstrate that the proposed R-JDRDL outperforms existing state-of-the-arts algorithms when used for image classification tasks.
AB - Dictionary leaning (DL) and dimensionality reduction (DR) are powerful tools to analyze high-dimensional noisy signals. This paper presents a proposal of a novel Riemannian joint dimensionality reduction and dictionary learning (R-JDRDL) on symmetric positive definite (SPD) manifolds for classification tasks. The joint learning considers the interaction between dimensionality reduction and dictionary learning procedures by connecting them into a unified framework. We exploit a Riemannian optimization framework for solving DL and DR problems jointly. Finally, we demonstrate that the proposed R-JDRDL outperforms existing state-of-the-arts algorithms when used for image classification tasks.
KW - Dictionary leaning
KW - Dimensionality reduction
KW - Riemannian manifold
KW - SPD matrix
UR - http://www.scopus.com/inward/record.url?scp=85059809359&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85059809359&partnerID=8YFLogxK
U2 - 10.23919/EUSIPCO.2018.8553200
DO - 10.23919/EUSIPCO.2018.8553200
M3 - Conference contribution
AN - SCOPUS:85059809359
T3 - European Signal Processing Conference
SP - 2010
EP - 2014
BT - 2018 26th European Signal Processing Conference, EUSIPCO 2018
PB - European Signal Processing Conference, EUSIPCO
T2 - 26th European Signal Processing Conference, EUSIPCO 2018
Y2 - 3 September 2018 through 7 September 2018
ER -