Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to main content

Advertisement

Springer Nature Link
Log in

Self-organizing mappings on the Grassmannian with applications to data analysis in high dimensions

  • WSOM 2017
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

We propose a method for extending Kohonen’s self-organizing mapping to the geometric framework of the Grassmannian. The resulting algorithm serves as a prototype of the extension of the SOM to the setting of abstract manifolds. The ingredients required for this are a means to measure distance between two points, and a method to move one point in the direction of another. In practice, the data are not required to have a representation in Euclidean space. We discuss in detail how a point on a Grassmannian is moved in the direction of another along a geodesic path. We demonstrate the implementation of the algorithm on several illustrative data sets, hyperspectral images and gene expression data sets.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

Notes

  1. Of course in this example the ordering of the points on the Grassmannian is available to us. However, in general we can determine a one-dimensional parameterization of a set of points on a Grassmannian that approximately passes through nearest neighbors.

References

  1. Absil P-A, Mahony R, Sepulchre R (2004) Riemannian geometry of Grassmann manifolds with a view on algorithmic computation. Acta Appl Math 80(2):199–220

    Article  MathSciNet  Google Scholar 

  2. Beveridge JR, Draper B, Chang J-M, Kirby M, Kley H, Peterson C (2008) Principal angles separate subject illumination spaces in YDB and CMU-PIE. IEEE Trans Pattern Anal Mach Intell 31:351–363

    Article  Google Scholar 

  3. Björck A, Golub GH (1973) Numerical methods for computing angles between linear subspaces. Math Comput 27(123):579–594

    Article  MathSciNet  Google Scholar 

  4. Chang J-M, Beveridge JR, Draper B, Kirby M, Kley H, Peterson C (2006) Illumination face spaces are idiosyncratic. In: IPCV’06, vol 2. CSREA Press, pp 390–396

  5. Chang J-M, Kirby M, Kley H, Peterson C, Beveridge JR, Draper B (2006) Examples of set-to-set pattern classification. In: Mathematics in signal processing conference digest, Royal Agricultural College, Cirencester, UK, The Insititute for Mathematics and its Applications, pp 102–105

  6. Chang J-M, Kirby M, Kley H, Peterson C, Draper B, Ross BJ (2007) Recognition of digital images of the human face at ultra low resolution via illumination spaces. In: Computer vision—ACCV 2007. Springer, pp 733–743

  7. Chang J-M, Kirby M, Peterson C (2007) Set-to-set face recognition under variations in pose and illumination. In: 2007 Biometrics symposium, Baltimore, MD

  8. Chepushtanova S, Gittins C, Kirby M (2014) Band selection in hyperspectral imagery using sparse support vector machines. In: Miguel V-R, Fred AK (eds) Algorithms and technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XX, volume 9088.  International Society for Optics and Photonics, 2014

  9. Chepushtanova S, Kirby M (2017) Sparse Grassmannian embeddings for hyperspectral data representation and classification. IEEE Geosci Remote Sens Lett 14(3):434–438

    Article  Google Scholar 

  10. Chepushtanova S, Kirby M, Peterson C, Ziegelmeier L (2015) An application of persistent homology on Grassmann manifolds for the detection of signals in hyperspectral imagery. In: Proceedings of the IEEE international geoscience and remote sensing symposium (IGARSS), Milan, Italy, 2015

  11. Conway JH, Hardin RH, Sloane NJA (1996) Packing lines, planes, etc.: packings in Grassmannian spaces. Exp Math 5(2):139–159

    Article  MathSciNet  Google Scholar 

  12. Draper B, Kirby M, Marks J, Marrinan T, Peterson C (2014) A flag representation for finite collections of subspaces of mixed dimensions. Linear Algebra Appl 451:15–32

    Article  MathSciNet  Google Scholar 

  13. Edelman A, Arias TA, Smith ST (1998) The geometry of algorithms with orthogonality constraints. SIAM J Matrix Anal Appl 20(2):303–353

    Article  MathSciNet  Google Scholar 

  14. Kaski S, Kangas J, Kohonen T (1998) Bibliography of self-organizing map (som) papers: 1981–1997. Neural Comput Surv 1(3&4):1–176

    Google Scholar 

  15. Kirby M, Peterson C (2017) Visualizing data sets on the grassmannian using self-organizing mappings. In: 2017 12th international workshop on self-organizing maps and learning vector quantization, clustering and data visualization (WSOM), pp 1–6

  16. Kohonen T (1982) Self-organized formation of topologically correct feature maps. Biol Cybern 43:59

    Article  MathSciNet  Google Scholar 

  17. Kohonen T (1990) The self-organizing map. Proc IEEE 78(9):1464–1480

    Article  Google Scholar 

  18. Kohonen T (1998) The self-organizing map. Neurocomputing 21(1):1–6

    Article  MathSciNet  Google Scholar 

  19. Kohonen T (2013) Essentials of the self-organizing map. Neural Netw 37:52–65

    Article  Google Scholar 

  20. Kohonen T, Oja E, Simula O, Visa A, Kangas J (1996) Engineering applications of the self-organizing map. Proc IEEE 84(10):1358–1384

    Article  Google Scholar 

  21. Landgrebe D (1992) AVIRIS NW Indiana’s Indian Pines 1992 data set

  22. Liu T-Y, Burke T, Park LP, Woods CW, Zaas AK, Ginsburg GS, Hero AO (2016) An individualized predictor of health and disease using paired reference and target samples. BMC Bioinf 17(1):47

    Article  Google Scholar 

  23. Marrinan T, Ross BJ, Draper B, Kirby M, Peterson C (2015) Flag manifolds for the characterization of geometric structure in large data sets. In: Numerical mathematics and advanced applications-ENUMATH 2013. Springer, pp 457–465

  24. Marrinan T, Draper B, Ross BJ, Kirby M, Peterson C (2014) Finding the subspace mean or median to fit your need. In: 2014 IEEE conference on computer vision and pattern recognition (CVPR). IEEE, pp 1082–1089

  25. Marrinan T, Ross BJ, Draper B, Kirby M, Peterson C (2016) Flag-based detection of weak gas signatures in long-wave infrared hyperspectral image sequences. In: Proceedings SPIE defense + security. International Society for Optics and Photonics, pp 98401N–98401N

  26. O’Hara S, Wang K, Slayden RA, Schenkel AR, Huber G, O’Hern CS, Shattuck MD, Kirby M (2013) Iterative feature removal yields highly discriminative pathways. BMC Genom 14(1):832

    Article  Google Scholar 

  27. Ritter H, Kohonen T (1989) Self-organizing semantic maps. Biol Cybern 61(4):241–254

    Article  Google Scholar 

  28. Rubins KH, Hensley LE, Wahl-Jensen V, Daddario DiCaprio KM, Young HA, Reed DS, Jahrling PB, Brown PO, Relman DA, Geisbert TW (2007) The temporal program of peripheral blood gene expression in the response of nonhuman primates to Ebola hemorrhagic fever. Genome Biol 8(8):R174

    Article  Google Scholar 

  29. Santamaria I, Scharf LL, Peterson C, Kirby M, Francos J (2016) An order fitting rule for optimal subspace averaging. In: Statistical signal processing workshop (SSP), 2016 IEEE. IEEE, pp 1–4

  30. Wang K, Langevin S, O’Hern C, Shattuck M, Ogle S, Forero A, Morrison J, Slayden R, Katze M, Kirby M (2016) Anomaly detection in host signaling pathways for the early prognosis of acute infection. PloS ONE 11:e0160919

    Article  Google Scholar 

Download references

Acknowledgements

This work is supported by the National Natural Science Foundation of China under Grant 61771178. We want to thank Shannon Stiverson for discovering the Reactome interferon signal pathway as discriminatory at 48–72 h. This paper is based on research partially supported by the National Science Foundation under Grants No. DMS-1513633, DMS-1322508, and CISE-1712788, as well as DARPA awards N66001-17-2-4020 and D17AP00004.

Author information

Authors and Affiliations

  1. Department of Mathematics, Colorado State University, Fort Collins, CO, USA

    Xiaofeng Ma, Michael Kirby & Chris Peterson

  2. Department of Mathematics and Statistics, Colorado State University, Fort Collins, CO, USA

    Louis Scharf

Authors
  1. Xiaofeng Ma
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Michael Kirby
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Chris Peterson
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Louis Scharf
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Michael Kirby.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ma, X., Kirby, M., Peterson, C. et al. Self-organizing mappings on the Grassmannian with applications to data analysis in high dimensions. Neural Comput & Applic 32, 18243–18254 (2020). https://doi.org/10.1007/s00521-019-04444-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-019-04444-x

Keywords

Search

Navigation