mahout-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Kun Yang (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (MAHOUT-1273) Single Pass Algorithm for Penalized Linear Regression with Cross Validation on MapReduce
Date Wed, 31 Jul 2013 22:11:50 GMT

    [ https://issues.apache.org/jira/browse/MAHOUT-1273?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13725770#comment-13725770
] 

Kun Yang commented on MAHOUT-1273:
----------------------------------

Another error message
Something broke! (Error 500)

It appears something broke when you tried to go to here. This is either a bug in Review Board
or a server configuration error. Please report this to your administrator.
                
> Single Pass Algorithm for Penalized Linear Regression with Cross Validation on MapReduce
> ----------------------------------------------------------------------------------------
>
>                 Key: MAHOUT-1273
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-1273
>             Project: Mahout
>          Issue Type: New Feature
>    Affects Versions: 0.9
>            Reporter: Kun Yang
>              Labels: documentation, features, patch, test
>             Fix For: 0.9
>
>         Attachments: Algorithm and Numeric Stability.pdf, java files.pdf, Manual and
Example.pdf, PenalizedLinear.pdf, PenalizedLinearRegression.patch
>
>   Original Estimate: 720h
>  Remaining Estimate: 720h
>
> Penalized linear regression such as Lasso, Elastic-net are widely used in machine learning,
but there are no very efficient scalable implementations on MapReduce.
> The published distributed algorithms for solving this problem is either iterative (which
is not good for MapReduce, see Steven Boyd's paper) or approximate (what if we need exact
solutions, see Paralleled stochastic gradient descent); another disadvantage of these algorithms
is that they can not do cross validation in the training phase, which requires a user-specified
penalty parameter in advance. 
> My ideas can train the model with cross validation in a single pass. They are based on
some simple observations.
> The core algorithm is a modified version of coordinate descent (see J. Freedman's paper).
They implemented a very efficient R package "glmnet", which is the de facto standard of penalized
regression.
> I have implemented the primitive version of this algorithm in Alpine Data Labs.  

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message