Mailing-List: contact dev-help@mahout.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@mahout.apache.org
Date: Wed, 31 Jul 2013 18:49:49 +0000 (UTC)
From: "Kun Yang (JIRA)" <jira@apache.org>
To: dev@mahout.apache.org
Message-ID: <JIRA.12656094.1372877268484.146372.1375296589022@arcas>
In-Reply-To: <JIRA.12656094.1372877268484@arcas>
References: <JIRA.12656094.1372877268484@arcas>
Subject: [jira] [Updated] (MAHOUT-1273) Single Pass Algorithm for Penalized
 Linear Regression with Cross Validation on MapReduce
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


     [ https://issues.apache.org/jira/browse/MAHOUT-1273?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Kun Yang updated MAHOUT-1273:
-----------------------------

    Attachment: java files.pdf
                Manual and Example.pdf
                Algorithm and Numeric Stability.pdf
    
> 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
>            Reporter: Kun Yang
>         Attachments: Algorithm and Numeric Stability.pdf, java files.pdf, Manual and Example.pdf, PenalizedLinear.pdf
>
>   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