[
https://issues.apache.org/jira/browse/MATH-1563?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17244748#comment-17244748
]
Gilles Sadowski commented on MATH-1563:
---------------------------------------
{quote}I would like [...]
{quote}
Thanks a lot for your offer.
As you may know (if you've been subscribed for a long time to the "dev" ML of
the "Commons" project), the "Commons Math" library is in dire need of
contributors in order to advance towards a consistent set of tools that are
maintained by developers who actually use them. To make a long story short:
Falling short of that goal entails inordinately hard maintenance and bugs that
linger for years (cf. the long list of them here) because each of the many
different parts of the library may require a specific expertise.
This is the case for the {{o.a.c.m.genetics}} package. In other such
occurrences, we much improved the situation (bugs-wise, features-wise,
maintenance-wise and performance-wise) by splitting the concerned functionality
off to a new component:
* [Commons RNG|http://commons.apache.org/proper/commons-rng/]
* [Commons Numbers|http://commons.apache.org/proper/commons-numbers/]
* [Commons Geometry|http://commons.apache.org/proper/commons-geometry/]
* [Commons Statistics
(distributions)|http://commons.apache.org/proper/commons-statistics/]
Thus, I think that we must assess whether the "genetic algorithms"
functionality has a reasonable future within "Apache Commons" (i.e. potential
users and contributors) while there exists other libraries that seem much more
advanced for any serious usage.
Of course, if you are an expert in that field, and if you are willing to spend
time to develop, and help maintain, a "competitive" alternative (in terms of
features and/or performance), then we should consider the creation of a new
component.
Please post a message on the "dev" ML, in order to get more opinions.
> Implementation of Adaptive Probability Generation Strategy for Genetic
> Algorithm
> --------------------------------------------------------------------------------
>
> Key: MATH-1563
> URL: https://issues.apache.org/jira/browse/MATH-1563
> Project: Commons Math
> Issue Type: Improvement
> Reporter: AVIJIT BASAK
> Priority: Major
>
> In Genetic Algorithm probability of crossover and mutation operation can be
> generated in an adaptive manner. Some experiment was done related to this and
> published in this article
> "https://www.ijcaonline.org/archives/volume175/number10/basak-2020-ijca-920572.pdf";.
> Currently Apache's API works on constant probability strategy. I would like
> to propose incorporation of rank based adaptive probability generation
> strategy as described in the mentioned article. This will improve the
> performance and robustness of the algorithm and would make this more suitable
> for use in higher dimensional problems like machine learning or deep learning.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
