[ 
https://issues.apache.org/jira/browse/MAHOUT-796?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13091545#comment-13091545
 ] 

Nathan Halko edited comment on MAHOUT-796 at 8/26/11 3:32 AM:
--------------------------------------------------------------

I checked Dmitriy's scheme today and it makes sense.  It accumulates Q' using 
the machinery already in place, QJob and BtJob

QR = Y = A\Omega
B0 = Q'A
B1 = (AB0')'A = B0A'A = (Q'AA')A = (Q_new)'A 

A'A should never be computed, only Z = A'AY where Y is dense and X=AY, Z=A'X 
avoiding the problem of scarce overlap and fill in.
 

      was (Author: nathanhalko):
    I checked Dmitriy's scheme today and it makes sense.  It accumulates Q* 
using the machinery already in place, QJob and BtJob

QR = Y = A\Omega
B0 = Q*A
B1 = (AB0*)*A = B0A*A = (Q*AA*)A = (Q_new)*A 

A'A should never be computed, only Z = A'AY where Y is dense and X=AY, Z=A'X 
avoiding the problem of scarce overlap and fill in.
 
  
> Modified power iterations in existing SSVD code
> -----------------------------------------------
>
>                 Key: MAHOUT-796
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-796
>             Project: Mahout
>          Issue Type: Improvement
>          Components: Math
>    Affects Versions: 0.5
>            Reporter: Dmitriy Lyubimov
>            Assignee: Dmitriy Lyubimov
>              Labels: SSVD
>             Fix For: 0.6
>
>
> Nathan Halko contacted me and pointed out importance of availability of power 
> iterations and their significant effect on accuracy of smaller eigenvalues 
> and noise attenuation. 
> Essentially, we would like to introduce yet another job parameter, q, that 
> governs amount of optional power iterations. The suggestion how to modify the 
> algorithm is outlined here : 
> https://github.com/dlyubimov/ssvd-lsi/wiki/Power-iterations-scratchpad .
> Note that it is different from original power iterations formula in the paper 
> in the sense that additional orthogonalization performed after each 
> iteration. Nathan points out that that improves errors in smaller eigenvalues 
> a lot (If i interpret it right). 

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Reply via email to