Principal component analysis: Difference between revisions

From Machinelearning
Line 5: Line 5:
* multiple ways to look at PCA:
* multiple ways to look at PCA:
** SVD on the covariance matrix (this is probably the same as one of the other interpretations)
** SVD on the covariance matrix (this is probably the same as one of the other interpretations)
** maximum variance (see Bishop)
** maximum variance (see Bishop). This one uses the Lagrange multiplier and [[derivative of a quadratic form]].
** minimum-error (see Bishop)
** minimum-error (see Bishop)
** the best linear compression-recovery of data to a lower dimension (see Shalev-Shwartz and Ben-David). Is this the same as minimum-error interpretation?
** the best linear compression-recovery of data to a lower dimension (see Shalev-Shwartz and Ben-David). Is this the same as minimum-error interpretation?

Revision as of 03:38, 14 July 2018

Questions/things to explain

  • Analogously to the covariance matrix one can define a correlation matrix. What happens if you run SVD on the correlation matrix?
  • multiple ways to look at PCA:
    • SVD on the covariance matrix (this is probably the same as one of the other interpretations)
    • maximum variance (see Bishop). This one uses the Lagrange multiplier and derivative of a quadratic form.
    • minimum-error (see Bishop)
    • the best linear compression-recovery of data to a lower dimension (see Shalev-Shwartz and Ben-David). Is this the same as minimum-error interpretation?