Abstract
We describe a new variant of the MBO scheme for solving the semi-supervised data classification problem on a weighted graph. The scheme is based on the minimization of the graph heat content energy. The resulting algorithms guarantee dissipation of the graph heat content energy for an extremely wide class of weight matrices. As a result, our method is both flexible and unconditionally stable. Experimental results on benchmark machine learning datasets show that our approach matches or exceeds the performance of current state-of-the-art variational methods while being considerably faster.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Alberti, G., Bellettini, G.: A non-local anisotropic model for phase transitions: asymptotic behavior of rescaled energies. Eur. J. Appl. Math. 9, 261–284 (1998)
Bresson, X., Chan, T., Tai, X., Szlam, A.: Multi-class trans- ductive learning based on l1 relaxations of cheeger cut and mumford-shah-potts model. J. Math. Imaging Vis. 49(1), 191–201 (2013)
Cheeger, J.: A lower bound for the smallest eigenvalue of the Laplacian. In: Problems in Analysis, pp. 195–199 (1970)
Coifman, R.R., Lafon, S., Lee, A.B., Maggioni, M., Nadler, B., Warner, F., Zucker, S.W.: Geometric diffusions as a tool for harmonic analysis and structure definition of data: diffusion maps. PNAS 102, 7426–7431 (2005)
Elsey, M., Esedo\(\bar{\text{g}}\)lu, S.: Threshold dynamics for anisotropic surface energies. Technical report, UM (2016). Under review
Esedo\(\bar{\text{ g }}\)lu, S., Otto, F.: Threshold dynamics for networks with arbitrary surface tensions. Commun. Pure Appl. Math. 68(5), 808–864 (2015)
Esedo\(\bar{\text{ g }}\)lu, S., Tsai, Y.-H.: Threshold dynamics for the piecewise constant Mumford-Shah functional. J. Comput. Phys. 211(1), 367–384 (2006)
Esedo\(\bar{\text{ g }}\)lu, S., Jacobs, M.: Convolution kernels, and stability of threshold dynamics methods. Technical report, University of Michigan (2016)
Garcia-Cardona, C., Merkurjev, E., Bertozzi, A.L., Flenner, A., Percus, A.G.: Multiclass data segmentation using diffuse interface methods on graphs. IEEE Trans. Pattern Anal. Mach. Intell. 36(8), 1600–1613 (2014)
Hein, M., Setzer, S.: Beyond spectral clustering - tight relaxations of balanced graph cuts. In: Advances in Neural Information Processing Systems 24 (NIPS) (2011)
Kaynak, C.: Methods of combining multiple classifiers and their applications to handwritten digit recognition. Master’s thesis, Institute of Graduate Studies in Science and Engineering, Bogazici University (1995)
Malik, J., Shi, J.: Normalized cuts and image segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 22(8), 888–905 (2000)
Merkurjev, E., Bertozzi, A., Chung, F.: A semi-supervised heat kernel pagerank MBO algorithm for data classification (2016, submitted)
Merriman, B., Bence, J.K., Osher, S.J.: Diffusion generated motion by mean curvature. In: Taylor, J. (ed.) Proceedings of the Computational Crystal Growers Workshop, pp. 73–83. AMS (1992)
Miranda, M., Pallara, D., Paronetto, F., Preunkert, M.: Short-time heat flow and functions of bounded variation in \({\mathbb{R}}^N\). Ann. Fac. Sci. Toulouse Math. 16(1), 125–145 (2007)
Mumford, D., Shah, J.: Optimal approximations by piecewise smooth functions and associated variational problems. Commun. Pure Appl. Math. 42, 577–685 (1989)
Nene, S.A., Nayar, S.K., Murase, H.: Columbia object image library (coil-100). Technical report, Columbia University (1996)
Zien, A., Chapelle, O., Scholkopf, B.: Semi-Supervised Learning. The MIT Press, Cambridge (2006)
Scherzer, O. (ed.): Handbook of Mathematical Methods in Imaging. Springer, Heidelberg (2011)
Yin, K., Tai, X.-Y., Osher, S.J.: An effective region force for some variational models for learning and clustering. Technical report, UCLA (2016)
Yu, S.X., Shi, J.: Multiclass spectral clustering. In: Ninth IEEE International Coference on Computer Vision, vol. 1, pp. 313–319, October 2003
Zelnik-Manor, L., Perona, P.: Self-tuning spectral clustering. In: Advances in Neural Information Processing Systems (2004)
Acknowledgments
The author is grateful to Selim Esedoḡlu for helpful comments and suggestions. The author was supported by NSF DMS-1317730.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Jacobs, M. (2017). A Fast MBO Scheme for Multiclass Data Classification. In: Lauze, F., Dong, Y., Dahl, A. (eds) Scale Space and Variational Methods in Computer Vision. SSVM 2017. Lecture Notes in Computer Science(), vol 10302. Springer, Cham. https://doi.org/10.1007/978-3-319-58771-4_27
Download citation
DOI: https://doi.org/10.1007/978-3-319-58771-4_27
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-58770-7
Online ISBN: 978-3-319-58771-4
eBook Packages: Computer ScienceComputer Science (R0)