Abstract
We propose a new class of Riemannian conjugate gradient (CG) methods, in which inverse retraction is used instead of vector transport for search direction construction. In existing methods, differentiated retraction is often used for vector transport to move the previous search direction to the current tangent space. However, a different perspective is adopted here, motivated by the fact that inverse retraction directly measures the displacement from the current to the previous points in terms of tangent vectors at the current point. The proposed algorithm is implemented with the Fletcher–Reeves and the Dai–Yuan formulae, respectively, and global convergence is established using modifications of the Riemannian Wolfe conditions. Computational details of the practical inverse retractions over the Stiefel and fixed-rank manifolds are discussed. Numerical results obtained for the Brockett cost function minimization problem, the joint diagonalization problem, and the low-rank matrix completion problem demonstrate the potential effectiveness of Riemannian CG with inverse retraction.
Similar content being viewed by others
References
Absil, P.-A., Baker, C. G., Gallivan, K. A.: Trust-region methods on Riemannian manifolds. Found. Comput. Math. 7, 303–330 (2007)
Absil, P.-A., Mahony, R., Sepulchre, R.: Optimization Algorithms on Matrix Manifolds. Princeton University Press, Princeton, NJ (2008)
Absil, P.-A., Malick, J.: Projection-like retractions on matrix manifolds. SIAM J. Optim. 22, 135–158 (2012)
Absil, P.-A., Oseledets, I. V.: Low-rank retractions: a survey and new results, Comput. Optim. Appl. 62, 5–29 (2015)
Al-Baali, M.: Descent property and global convergence of the Fletcher-Reeves method with inexact line search. IMA J. Numer. Anal. 5, 121–124 (1985)
Bonnabel, S.: Stochastic gradient descent on Riemannian manifolds. IEEE Trans. Autom. Control 58, 2217–2229 (2013)
Boumal, N., Mishra, B., Absil, P.-A., Sepulchre, R.: Manopt, a Matlab toolbox for optimization on manifolds. J. Mach. Learn. Res. 15, 1455–1459 (2014)
Cardoso, J.F., Souloumiac, A.: Blind beamforming for non-Gaussian signals. IEEE Proceedings F-Radar and Signal Processing 140, 362–370 (1993)
Dai, Y., Yuan, Y.: A nonlinear conjugate gradient method with a strong global convergence property. SIAM J, Optim. 10, 177–182 (1999)
do Carmo, M. P.: Riemannian Geometry. Translated from the second Portuguese edition by Francis Flaherty. Mathematics: Theory & Applications. Birkh\(\ddot{\rm a}\)user Boston Inc., Boston, MA (1992)
Edelman, A., Arias, T. A., Smith, S. T.: The geometry of algorithms with orthogonality constraints. SIAM J. Matrix Anal. Appl. 20, 303–353 (1998)
Fletcher, R., Reeves, C. M.: Function minimization by conjugate gradients. Comput. J. 7, 149–154 (1964)
Gallivan, K. A., Qi, C., Absil, P.-A.: A Riemannian Dennis-Moré condition. In: Berry, M.W., Gallivan, K.A., Gallopoulos, E., Grama, A., Philippe, B., Saad, Y., Saied, F. (eds.) High-Performance Scientific Computing, Springer, London, pp. 281–293 (2012)
Grohs, P., Hosseini, S.: \(\varepsilon\)-subgradient algorithms for locally Lipschitz functions on Riemannian manifolds. Adv. Comput. Math. 42, 333–360 (2016)
Hosseini, S., Huang, W., Yousefpour, R.: Line search algorithms for locally Lipschitz functions on Riemannian manifolds. SIAM J. Optim. 28, 596–619 (2018)
Hosseini, S., Uschmajew, A.: A Riemannian gradient sampling algorithm for nonsmooth optimization on manifolds. SIAM J. Optim. 27, 173–189 (2017)
Hosseini, S., Uschmajew, A.: A gradient sampling method on algebraic varieties and application to nonsmooth low-rank optimization. SIAM J. Optim. 29, 2853–2880 (2019)
Hu, J., Liu, X., Wen, Z., Yuan, Y.: A brief introduction to manifold optimization. J. Oper. Res. Soc. China, published online, https://doi.org/10.1007/s40305-020-00295-9
Huang, W.: Optimization algorithms on Riemannian manifolds with applications. Ph.D. thesis, Department of Mathematics, Florida State University (2013)
Huang, W., Absil, P.-A., Gallivan, K. A.: A symmetric rank-one trust-region method. Math. Program. 150, 179–216 (2015)
Huang, W., Absil, P.-A., Gallivan, K. A.: A Riemannian BFGS method without differentiated retraction for nonconvex optimization problems. SIAM J. Optim. 28, 470–495 (2018)
Huang, W., Gallivan, K. A., Absil, P.-A.: A Broyden class of quasi-Newton methods for Riemannian optimization. SIAM J. Optim. 25, 1660–1685 (2015)
Huang, W., Wei, K.: Extending FISTA to Riemannian optimization for sparse PCA. arXiv:1909.05485v1 (2019)
Huang, W., Wei, K.: Riemannnian proximal gradient methods. arXiv:1909.06065v1 (2019)
Jiang, B., Ma, S., Man-Cho So, A., Zhang, S.: Vector transport-free SVRG with general retraction for Riemannian optimization: complexity analysis and practical implementation. arXiv:1705.09059v1 (2017)
Kaneko, T., Fiori, S., Tanaka, T.: Empirical arithmetic averaging over the compact Stiefel manifold. IEEE Transactions on Signal Processing 61, 883–894 (2013)
Koch, O., Lubich, C.: Dynamical low-rank approximation. SIAM J. Matrix Anal. Appl. 29, 434–454 (2007)
Lee, J.M.: Introduction to Smooth Manifolds, 2nd edn. Springer, New York, NY (2012)
Nocedal, J., Wright, S. J.: Numerical Optimization, 2nd ed., Springer, New York, NY (2006)
Qi, C., Gallivan, K. A., Absil, P.-A.: Riemannian BFGS algorithm with applications. In: Recent Advances in Optimization and its Applications in Engineering, Springer-Verlag Berlin, Heidelberg, pp. 183–192 (2010)
Ring, W., Wirth, B.: Optimization methods on Riemannian manifolds and their application to shape space. SIAM J. Optim. 22, 596–627 (2012)
Rockafellar, R. T., Wets, J.-B.: Variational Analysis. Springer, Berlin, Heidelberg (1998)
Sato, H.: A Dai-Yuan-type Riemannian conjugate gradient method with the weak Wolfe conditions. Comput. Optim. Appl. 64, 101–118 (2016)
Sato, H., Iwai, T.: A new, globally convergent Riemannian conjugate gradient method. Optimization 64, 1011–1031 (2015)
Sato, H., Kasai, H., Mishra, B.: Riemannian stochastic variance reduced gradient algorithm with retraction and vector transport. SIAM J. Optim. 29, 1444–1472 (2019)
Siegel, J. W.: Accelerated optimization with orthogonality constraints. arXiv:1903.05204v3 (2019)
Sun, W., Yuan, Y.: Optimization Theory and Methods: Nonlinear Programming. Springer, Boston, MA (2006)
Theis, F.J., Cason, T.P., Absil, P.-A.: Soft dimension reduction for ICA by joint diagonalization on the Stiefel manifold. In: Adali, T., Jutten, C., Romano, J.M.T., Barros, A. (eds.) Independent Component Analysis and Signal Separation. Lecture Notes in Computer Science, vol. 5441, pp. 354–361. Springer, Berlin (2009)
Tripuraneni, N., Flammarion, N., Bach, F., Jordan, M. I.: Averaging stochastic gradient descent on Riemannian manifolds. PMLR 75, 1–38 (2018)
Vandereycken, B.: Low-rank matrix completion by Riemannian optimization. SIAM J. Optim. 23, 1214–1236 (2013)
Wen, Z., Yin, W.: A feasible method for optimization with orthogonality constraints. Math. Program. 142, 397–434 (2013)
Zhang, H., Reddi, S. J., Sra, S.: Fast stochastic optimization on Riemannian manifolds. arXiv:1605.07147v1 (2016)
Zhang, H., Sra, S.: First-order methods for geodesically convex optimization. In: JMLR: Workshop and Conference Proceedings vol. 49, pp. 1–22 (2016)
Zhang, H., Sra, S.: An estimate sequence for geodesically convex optimization. PMLR 75, 1–21 (2018)
Acknowledgements
X. Zhu is supported by National Natural Science Foundation of China Grant Number 11601317 and H. Sato is supported by JSPS KAKENHI Grant Number JP20K14359. The authors are grateful to the coordinating editor and two anonymous referees for their valuable comments and suggestions.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Zhu, X., Sato, H. Riemannian conjugate gradient methods with inverse retraction. Comput Optim Appl 77, 779–810 (2020). https://doi.org/10.1007/s10589-020-00219-6
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10589-020-00219-6