[
https://issues.apache.org/jira/browse/FLINK-1979?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15315212#comment-15315212
]
ASF GitHub Bot commented on FLINK-1979:
---------------------------------------
Github user skavulya commented on the issue:
https://github.com/apache/flink/pull/1985
@chiwanpark Decoupling the gradient descent step is complicated for L1
regularization because we are using the proximal gradient method that applies
soft thresholding after executing the gradient descent step. I left the
regularization penalty as-is. I am thinking of adding an additional method that
adds the regularization penalty to gradient without the gradient descent step
but I will do it in the L-BFGS PR instead.
> Implement Loss Functions
> ------------------------
>
> Key: FLINK-1979
> URL: https://issues.apache.org/jira/browse/FLINK-1979
> Project: Flink
> Issue Type: Improvement
> Components: Machine Learning Library
> Reporter: Johannes Günther
> Assignee: Johannes Günther
> Priority: Minor
> Labels: ML
>
> For convex optimization problems, optimizer methods like SGD rely on a
> pluggable implementation of a loss function and its first derivative.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
