Publications

 

Published peer-reviewed articles

  1. Virta J., Nordhausen K. (2017): Blind source separation for nonstationary tensor-valued time series, to appear in “2017 IEEE 27th International Workshop on Machine Learning for Signal Processing (MLSP)”.
  2. Virta J., Nordhausen K. (2017): Blind source separation of tensor-valued time series, Signal Processing, Volume 141, December 2017, p. 204 – 216, preprint in arXiv, temporary free access.
  3. Nordhausen K., Oja H., Tyler D.E., Virta J. (2017): Asymptotic and bootstrap tests for the dimension of the non-Gaussian subspace, IEEE Signal Processing Letters, Volume 24, Issue 6, June 2017, p. 887 – 891, preprint in arXiv.
  4. Virta J., Nordhausen K. (2017): On the optimal non-linearities for Gaussian mixtures in FastICA, in the proceedings of “International Conference on Latent Variable Analysis and Signal Separation”, p. 427 – 437.
  5. Miettinen J., Nordhausen K., Oja H., Taskinen S. and Virta J. (2017): The squared symmetric FastICA estimator, Signal Processing, Volume 131, February 2017, p. 402 – 411, preprint in arXiv.
  6. Virta J., Taskinen S. and Nordhausen K. (2016): Applying fully tensorial ICA to fMRI data, in the proceedings of “2016 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)”.
  7. Virta J. (2015): One-step M-estimates of scatter and the independence property, Statistics & Probability Letters, Volume 110, March 2016, p. 133 – 136.

Submitted manuscripts

  • Virta J., Nordhausen K. and Oja H. (2016): Projection pursuit for non-Gaussian independent components, submitted, arXiv link.
  • Virta J., Li B., Nordhausen K. and Oja H. (2016): JADE for tensor-valued observations, submitted, arXiv link.
  • Virta J., Li B., Nordhausen K. and Oja H. (2016): Independent component analysis for tensor-valued data, submitted, arXiv link.

Manuscripts under preparation

  • Virta J., Li B., Oja H. and Nordhausen K. (2017): Independent component analysis for multivariate functional data.
  • (2017): Projection pursuit for functional data.
  • Virta J., Nordhausen K. and Oja H. (2015): Solving the independent component problem with cumulant matrices.
  • Virta J., Li B., Nordhausen K. and Oja H. (2015): Projection pursuit-based independent component analysis for tensor data.

Other manuscripts

  • Virta J., Nordhausen K. and Oja H. (2015): Joint use of third and fourth cumulants in independent component analysis, preprint in arXiv.

Software

  • Nordhausen K., Oja H., Tyler D.E., Virta J. (2016): R-package ICtest: Estimating and Testing the Number of Interesting Components in Linear Dimension Reduction, available from CRAN.
  • Virta J., Li B., Nordhausen K., Oja H. (2016): R-package tensorBSS: Blind Source Separation Methods for Tensor-Valued Observations, available from CRAN.

Theses

  • Virta J. (2014): Some Tools for Linear Dimension Reduction, Master’s thesis, University of Turku, pdf.
  • Virta J. (2012): Pääkomponenttianalyysi (Principal Component Analysis), Bachelor’s thesis, University of Turku.