You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
You haven't stated the intended purpose of the poster.
I definitely support a poster that publicizes the achievements of the team behind an exceptional and sustained effort over the last 10+ years. I hope that it increases the number of researchers who use it as a standard. (Petty, perhaps, but I almost have an instant mistrust of many papers that do not use COCO when they could do so easily!)
My only major criticism of the present state of COCO is that it is still quite difficult to add more functions to the standard single objective suite, and then to routinely produce the same high quality post-processed results that are currently available.
That single improvement would go a long way to making COCO almost a de facto "industry standard" benchmark suite and system. But that's just me - I am very biased in your favour! :)
Thanks for your feedback. The poster could serve as a visual tool when presenting COCO to a person who hasn't used it before. Its purpose should therefore be to show COCO's capabilities and ease of use*. Not sure we want to show how much work has been put into it ...
* Your point about difficulty of adding new problems (or problems suits) is of course valid. However, from the perspective of someone who just wants to benchmark their algorithm on a set of existing problems, it is easy to use.
@nikohansen, @brockho and I have been brainstorming the idea to make a poster about COCO. It could/should contain the following elements:
The text was updated successfully, but these errors were encountered: