taskGneural Network - Tasks: task #14135, Support direct partial derivative...

 
 

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

task #14135: Support direct partial derivative functions.

Submitter:  Ray Dillinger <rayd>
Submitted:  Sat 27 Aug 2016 01:09:54 AM UTC
   
 
Should Start On:  Fri 26 Aug 2016 07:00:00 AM UTC Should be Finished on:  Mon 26 Sep 2016 07:00:00 AM UTC
Category:  None Priority:  5 - Normal
Status:  None Privacy:  Public
Assigned to:  None Percent Complete:  0%
Open/Closed:  Open Effort:  0.00

Sat 27 Aug 2016 01:09:54 AM UTC, original submission:  


The method of computing partial derivatives currently expressed in the code is inaccurate (the interval is sufficiently small that shallow gradients will sometimes be rounded to zero, and if made larger it will sometimes span topologically important features in the fitness space). 

It is also inefficient (requires multiple calls to the activation function). 

Both things can be fixed by using explicit derivative functions which need be called only once.

Ray Dillinger <rayd>
Group Member

 

(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 rayd (Submitted the item)
  • -email is unavailable- added by rayd
  •  

    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 2 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2016-08-27 rayd Should be Finished on2016-08-26 2016-09-26
    2016-08-27 rayd Carbon-Copy- Added -email is unavailable-

    Back to the top

    Powered by Savane 3.13-02a9.
    Corresponding source code