taskGNU Astronomy Utilities - Tasks: task #15080, Machine learning in Gnuastro

 
 

You are not allowed to post comments on this tracker with your current authentication level.

task #15080: Machine learning in Gnuastro

Submitter:  Mohammad Akhlaghi <makhlaghi>
Submitted:  Tue 30 Oct 2018 01:20:50 PM UTC
   
 
Should Start On:  Mon 29 Oct 2018 11:00:00 PM UTC Should be Finished on:  Mon 29 Oct 2018 11:00:00 PM UTC
Category:  Libraries Priority:  5 - Normal
Item Group:  None Status:  In Progress
Privacy:  Public Assigned to:  None
Percent Complete:  0% Open/Closed:  Open
Effort:  0.00

Fri 27 Mar 2020 09:18:36 PM UTC, comment #2: 

This would have been a good task for a full GSoC proposal, but probably its too un-related to the original proposal (a Julia wrapper for Gnuastro library) to be included with it in one proposal. If you have extra time in the planning, I recommend to allocate more time to the low-level improvements of the Gnuastro library (to be more wrapper-friendly) and adding library interfaces to the existing programs (as described in comment 15 of task #13786).

Mohammad Akhlaghi <makhlaghi>
Group administrator
Fri 27 Mar 2020 08:15:53 PM UTC, comment #1: 

Is there any machine learning task or ideas available to be done during GSoC ? I am drawing near the end of my proposal and have figured out I have 30-40 days in spare.

Kartik Ohri <lucifer13>
Tue 30 Oct 2018 01:20:50 PM UTC, original submission:  

Machine learning libraries are increasingly being used in astronomy, or scientific applications in general. While it does have many caveats that must not be forgotten (see P.S.), it does also have benefits which can be useful. For example see arXiv:1810.10098 in task #14980.

I am starting this task for a general review of possible libraries (or generally, existing infra-structure) that we may be able to use in Gnuastro to enable Machine learning applications within Gnuastro's programs and libraries. Here the best summary list of libraries in various languages that I have found so far:

https://github.com/josephmisiti/awesome-machine-learning

We don't have to only look at those written in C. We can also link with libraries originally written in other languages, if they have a C interface. For example Torch is written in Lua, but they claim has an "amazing interface to C, via LuaJIT".

P.S. For a nice general discussion of Machine learning caveats see (link to YouTube) this talk by Ali Rahimi upon receiving the "Test-of-time" award at NIPS (Neural Information Processing Systems) 2017 conference. In this talk he compares Machine learning to alchemy. A (Facebook link) rebuttable was also published by Yann LeCun afterwards.

Mohammad Akhlaghi <makhlaghi>
Group administrator

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

No files currently attached

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -email is unavailable- added by pedram
  • -email is unavailable- added by makhlaghi (Submitted the item)
  •  

    There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.

     

    Follow 5 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2020-10-30 pedram Carbon-Copy- Added pedram
    2020-03-27 makhlaghi Carbon-CopyRemoved -email is unavailable- -
    2020-03-27 makhlaghi Carbon-CopyRemoved -email is unavailable- -
    2018-10-30 makhlaghi Carbon-Copy- Added -email is unavailable-
        Carbon-Copy- Added -email is unavailable-

    Back to the top

    Powered by Savane 3.13-4448.
    Corresponding source code