TY - GEN
T1 - A projection method for optimization problems on the Stiefel manifold
AU - Dalmau-Cedeño, Oscar
AU - Oviedo, Harry
N1 - Publisher Copyright:
© Springer International Publishing AG 2017.
PY - 2017
Y1 - 2017
N2 - In this paper we propose a feasible method based on projections using a curvilinear search for solving optimization problems with orthogonality constraints. Our algorithm computes the SVD decomposition in each iteration in order to preserve feasibility. Additionally, we present some convergence results. Finally, we perform numerical experiments with simulated problems; and analyze the performance of the proposed methods compared with state-of-the-art algorithms.
AB - In this paper we propose a feasible method based on projections using a curvilinear search for solving optimization problems with orthogonality constraints. Our algorithm computes the SVD decomposition in each iteration in order to preserve feasibility. Additionally, we present some convergence results. Finally, we perform numerical experiments with simulated problems; and analyze the performance of the proposed methods compared with state-of-the-art algorithms.
KW - Constrained optimization
KW - Non-monotone algorithm
KW - Optimization on manifolds
KW - Orthogonality constraints
KW - Stiefel manifold
UR - http://www.scopus.com/inward/record.url?scp=85021203928&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-59226-8_9
DO - 10.1007/978-3-319-59226-8_9
M3 - Conference contribution
AN - SCOPUS:85021203928
SN - 9783319592251
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 84
EP - 93
BT - Pattern Recognition - 9th Mexican Conference, MCPR 2017, Proceedings
A2 - Carrasco-Ochoa, Jesus Ariel
A2 - Martinez-Trinidad, Jose Francisco
A2 - Olvera-Lopez, Jose Arturo
PB - Springer Verlag
T2 - 9th Mexican Conference on Pattern Recognition, MCPR 2017
Y2 - 21 June 2017 through 24 June 2017
ER -