Closed-form geodesics and trust-region method to calculate Riemannian logarithms on Stiefel and its quotient manifolds (2103.13327v1)
Abstract: We provide two closed-form geodesic formulas for a family of metrics on Stiefel manifold, parameterized by two positive numbers, having both the embedded and canonical metrics as special cases. The closed-form formulas allow us to compute geodesics by matrix exponential in reduced dimension for low-rank manifolds. Combining with the use of Fr{\'e}chet derivatives to compute the gradient of the square Frobenius distance between a geodesic ending point to a given point on the manifold, we show the logarithm map and geodesic distance between two endpoints on the manifold could be computed by {\it minimizing} this square distance by a {\it trust-region} solver. This leads to a new framework to compute the geodesic distance for manifolds with known geodesic formula but no closed-form logarithm map. We show the approach works well for Stiefel as well as flag manifolds. The logarithm map could be used to compute the Riemannian center of mass for these manifolds equipped with the above metrics. We also deduce simple trigonometric formulas for the Riemannian exponential and logarithm maps on the Grassmann manifold.
- In: Proc. IEEE Conf. Comput. Vis. Pattern Recognit., pp. 1–8 (2008). DOI 10.1109/CVPR.2008.4587733
- Neurocomput. 67, 106–135 (2005). DOI 10.1016/j.neucom.2004.11.035
- Ann. Statist. 47(1), 415–438 (2019). DOI 10.1214/18-AOS1692
- SIAM J. Matrix Anal. Appl. 20(2), 303–353 (1999). DOI 10.1137/S0895479895290954
- Karcher, H.: Riemannian center of mass and mollifier smoothing. Comm. Pure and Applied Math. pp. 509–541 (1977). DOI 10.1002/cpa.3160300502
- Rentmeesters, Q.: Algorithms for data fitting on some common homogeneous spaces. Ph.D. thesis, Université Catholique de Louvain, Louvain, Belgium (2013). URL http://hdl.handle.net/2078.1/132587
- Zimmermann, R.: A matrix-algebraic algorithm for the Riemannian logarithm on the Stiefel manifold under the canonical metric. SIAM J. Matrix Anal. Appl. 38(2), 322–342 (2017). DOI 10.1137/16M1074485
- Bryner, D.: Endpoint geodesics on the Stiefel manifold embedded in Euclidean space. SIAM J. Matrix Anal. Appl. 38(4), 1139–1159 (2017). DOI 10.1137/16M1103099
- SIAM J. Imaging Sci. 4, 109–145 (2011). DOI doi.org/10.1137/090781139
- Journal of Geometric Mechanics (2020). DOI 10.3934/jgm.2020031
- Nguyen, D.: Operator-valued formulas for Riemannian gradient and Hessian and families of tractable metrics in optimization and machine learning. arXiv:2009.10159
- Springer (2016)
- SIAM J. Matrix Anal. Appl. 30(4), 1639–1657 (2009). DOI 10.1137/080716426
- Mathias, R.: A chain rule for matrix functions and applications. SIAM J. Matrix Anal. Appl. 17, 610–620 (1996). DOI 10.1137/S0895479895283409
- Adv. Appl. Math. 16, 321––375 (1995). DOI 10.1006/aama.1995.1017
- Springer, London, UK. (1994)
- In: J. Rosca, D. Erdogmus, J.C. Príncipe, S. Haykin (eds.) Independent Component Analysis and Blind Signal Separation, pp. 295–302. Springer Berlin Heidelberg (2006). DOI 10.1007/11679363˙37
- arXiv:1907.00949
- Springer International Publishing (2020)
- Linear Algebra and its Applications 466, 83–101 (2015). DOI 10.1016/j.laa.2014.10.003
- arXiv:2011.13699
- Springer (2008)
- Nguyen, D.: Project ManNullRange. https://github.com/dnguyend/ManNullRange (2021)