Publications

Published Papers:

1. Yoshitatsu Matsuda and Kazunori Yamaguchi (2012) .
   Global Mapping Analysis: Stochastic Gradient Algorithm in Multidimensional Scaling.
   IEICE TRANSACTIONS on Information and Systems, E95-D (2), 596-603.
   

2. Yoshitatsu Matsuda and Kazunori Yamaguchi (2011) .
   An Adaptive Threshold in Joint Approximate Diagonalization by Assuming Exponentially Distributed Errors.
   Neurocomputing, 74, 1994-2001.
   

3. Yoshitatsu Matsuda and Kazunori Yamaguchi (2009) .
   Linear Multilayer ICA Using Adaptive PCA.
   Neural Processing Letters, 30 (2), 133-144.
   

4. Yoshitatsu Matsuda and Kazunori Yamaguchi (2007) .
   Linear Multilayer ICA Generating Hierarchical Edge Detectors.
   Neural Computation, 19 (1), 218-230.
   

5. Yoshitatsu Matsuda and Kazunori Yamaguchi (2005) .
   An efficient MDS-based topographic mapping algorithm.
   Neurocomputing, 64, 285-299.
   

6. Yoshitatsu Matsuda and Kazunori Yamaguchi (2001) .
   Global mapping analysis: Stochastic approximation for multidimensional scaling.
   International Journal of Neural Systems, 11 (5), 419-426.
   

Proceedings of International Conference:

1. Yoshitatsu Matsuda and Kazunori Yamaguchi (2012) .
   A Robust Objective Function of Joint Approximate Diagonalization. ICANN2012, Lausanne, Switzerland, LNCS 7553, Springer-Verlag, 205-212.
   

2. Yoshitatsu Matsuda and Kazunori Yamaguchi (2011) .
   An Information Theoretic Approach to Joint Approximate Diagonalization. ICONIP2011, Shanghai, China, LNCS 7062, Springer-Verlag, 20-27.
   

3. Yoshitatsu Matsuda and Kazunori Yamaguchi (2010) .
   Partial Extraction of Edge Filters by Cumulant-based ICA. ICONIP2010, Sydney, Australia, LNCS 6444, Springer-Verlag, 633-640.
   

4. Yoshitatsu Matsuda and Kazunori Yamaguchi (2010) .
   A Probabilistic Framework for Joint Approximate Diagonalization. Proceedings of IJCNN2010, Barcelona, Spain, 364-369.
   

5. Takayuki Sekiya, Yoshitatsu Matsuda, and Kazunori Yamaguchi (2010) .
   Analysis of Computer Science Related Curriculum on LDA and Isomap. ITiCSE2010, Ankara, Turkey, 48-52.

6. Takayuki Sekiya, Yoshitatsu Matsuda, and Kazunori Yamaguchi (2010) .
   Development of a curriculum analysis tool. ITHET2010, Cappadocia, Turkey, 413-418.

7. Takayuki Sekiya, Yoshitatsu Matsuda, and Kazunori Yamaguchi (2009) .
   Analysis of Curriculum Structure Based on LDA. International Conference on Education and Information Technology, USA, 561-566.

8. Yoshitatsu Matsuda and Kazunori Yamaguchi (2009) .
   An Adaptive Threshold in Joint Approximate Diagonalization by the Information Criterion. ICONIP2009, Bangkok, Thailand, LNCS 5863, Springer-Verlag, 204-211.
   

9. Yoshitatsu Matsuda and Kazunori Yamaguchi (2009) .
   Joint Approximate Diagonalization Utilizing AIC-based Decision in the Jacobi Method. ICANN2009, Limassol, Cyprus, LNCS 5769, Springer-Verlag, 135-144.
   

10. Yoshitatsu Matsuda and Kazunori Yamaguchi (2009) .
    A Simple Overcomplete ICA Algorithm by Non-Orthogonal Pair Optimizations. Proceedings of IJCNN2009, Atlanta, Georgia, USA, 2027-2031.
    

11. Takayuki Sekiya, Yoshitatsu Matsuda, and Kazunori Yamaguchi (2008) .
    Systematization of Course Syllabi Based on LDA. SCIS&ISIS2008, Nagoya, Japan, 166-171.

12. Yoshitatsu Matsuda and Kazunori Yamaguchi (2008) .
    An Overcomplete ICA Algorithm by InfoMax and InfoMin. ICANN2008, Prague, Czech, LNCS 5163, Springer-Verlag, 136-144.
    

13. Yoshitatsu Matsuda and Kazunori Yamaguchi (2008) .
    A Connection-limited Neural Network by Infomax and Infomin. Proceedings of IJCNN2008, Hong Kong, 2531-2537.
    

14. Yoshitatsu Matsuda and Kazunori Yamaguchi (2007) .
    Extraction of Approximate Independent Components from Large Natural Scenes. ICONIP2007, Kitakyushu, Japan, LNCS 4984, Springer-Verlag, 635-642.
    

15. Yoshitatsu Matsuda and Kazunori Yamaguchi (2006) .
    A Fixed-Point Algorithm of Topographic ICA. ICANN2006, Athens, Greece, LNCS 4132, Springer-Verlag, 587-594.
    

16. Yoshitatsu Matsuda and Kazunori Yamaguchi (2006) .
    The InfoMin Principle for ICA and Topographic Mappings. ICA2006, Charleston, SC, USA, LNCS 3889, Springer-Verlag, 958-965.
    

17. Yoshitatsu Matsuda and Kazunori Yamaguchi (2005) .
    An efficient MDS algorithm for the analysis of massive document collections. KES2005, Melbourne, Australia, LNAI 3682, Springer-Verlag, 1015-1021.
    

18. Yoshitatsu Matsuda and Kazunori Yamaguchi (2005) .
    Linear multilayer independent component analysis for large natural scenes. NIPS2004, Vancouver, Canada, Advances in Neural Information Processing Systems 17, MIT Press, 897-904.
    

19. Yoshitatsu Matsuda and Kazunori Yamaguchi (2004) .
    Linear multilayer independent component analysis using stochastic gradient algorithm. ICA2004, Granada, Spain, LNCS 3195, Springer-Verlag, 303-310.
    

20. Yoshitatsu Matsuda and Kazunori Yamaguchi (2004) .
    Semi-invariant function of Jacobi algorithm in independent component analysis. Proceedings of 2004 IEEE International Joint Conference on Neural Networks (IJCNN2004), Budapest, Hungary, 2147-2151.
    

21. Yoshitatsu Matsuda and Kazunori Yamaguchi (2003) .
    The InfoMin criterion: An information theoretic unifying objective function for topographic mappings.
    Artificial Neural Network and Neural Information Processing - ICANN/ICONIP 2003, Istanbul, Turkey, LNCS 2714, Springer-Verlag, 401-408.
    

22. Yoshitatsu Matsuda and Kazunori Yamaguchi (2003) .
    Linear multilayer ICA algorithm integrating small local modules.
    Proceedings of ICA2003, Nara, Japan, 403-408.
    

23. Yoshitatsu Matsuda and Kazunori Yamaguchi (2001) .
    The InfoMin principle: A unifying information-based criterion for forming topographic mappings.
    Proceedings of ICONIP2001, Shanghai, China, 14-19.
    

24. Yoshitatsu Matsuda and Kazunori Yamaguchi (2001) .
    Global mapping analysis: Stochastic gradient algorithm in SSTRESS and classical MDS stress.
    Proceedings of ICONIP2001, Shanghai, China, 102-107.
    

Ph.D. Thesis:

1. Yoshitatsu Matsuda (2002) .
   Model and applications of the global topographic mapping formation.