[Computer-go] Evaluating improvements differently
dailey.don at gmail.com
Thu Apr 7 20:33:52 PDT 2011
On Thu, Apr 7, 2011 at 11:23 PM, terry mcintyre <terrymcintyre at yahoo.com>wrote:
> When training a shot putter, one does not merely practice the entire
> activity; one looks for weaknesses and devises strategies to strengthen the
> athlete in those areas. These strategies might include time on the weight
This is how you train, but it's not how you measure the athletes
performance in the shot put.
> We'll probably need a variety of ways to nudge programs to the next higher
> level of competition; a variety of tests which focus on specific things, and
> which encourage the programmers to think more deeply about certain aspects.
It all sounds great on paper, but so many ideas that don't pan out do too.
There are tests I use but the are very specific and primarily for
debugging. For go it would be useful to have a thousand ladder positions
and so on. But in the end you will notice that you have to play games to
see how anything like this translates to actual strength.
> Terry McIntyre <terrymcintyre at yahoo.com>
> Unix/Linux Systems Administration
> Taking time to do it right saves having to do it twice.
> Computer-go mailing list
> Computer-go at dvandva.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Computer-go