spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Xusen Yin (JIRA)" <j...@apache.org>
Subject [jira] [Created] (SPARK-1585) Not robust Lasso causes Infinity on weights and losses
Date Wed, 23 Apr 2014 09:46:17 GMT
Xusen Yin created SPARK-1585:
--------------------------------

             Summary: Not robust Lasso causes Infinity on weights and losses
                 Key: SPARK-1585
                 URL: https://issues.apache.org/jira/browse/SPARK-1585
             Project: Spark
          Issue Type: Bug
          Components: MLlib
    Affects Versions: 0.9.1
            Reporter: Xusen Yin
            Assignee: Xusen Yin
             Fix For: 1.0.0


Lasso uses LeastSquaresGradient and L1Updater, but 

diff = brzWeights.dot(brzData) - label

in LeastSquaresGradient would cause too big diff, then will affect the L1Updater, which increases
weights exponentially. Small shrinkage value cannot lasso weights back to zero then. Finally,
the weights and losses reach Infinity.

For example, data = (0.5 repeats 10k times), weights = (0.6 repeats 10k times), then data.dot(weights)
approximates 300+, the diff will be 300. Then L1Updater sets weights to approximate 300. In
the next iteration, the weights will be set to approximate 30000, and so on.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Mime
View raw message