[Computer-go] 50k-100k patterns

Mark Boon tesujisoftware at gmail.com
Fri Aug 3 13:20:32 PDT 2012


Interesting article. I've always been a little bit uncomfortable with
the concept of 'intuition'. It's probably nothing more than
pattern-matching on a level we just haven't seen yet in computers,
rather than the mythical quality most people take it to be.

I take issue with part of the article though:

"The first phase of the research showed that people's knowledge
undergoes dramatic reorganisation when they move from amateur to
professional rank.

“The change takes place not just in the areas of ‘deep strategy’,
where one would expect the big gains to be, but also there is a
radical reorganisation at the perceptual level,” Professor Bossomaier
said.

“It's a bit like acquiring a good accent in a foreign language. At
quite a young age, the sounds of one's first language get set and are
very difficult to change later."

I don't think the dramatic reorganisation happens when moving to
professional rank, and it's contradictory to the lines just below
about acquiring a language at young age. People who acquire a certain
skill at a very young age have a dramatically different organization
at the perceptual level, which does NOT change when they move on to
professional level. That's why you will not see real top Go players
who didn't learn the game between ages 4 and 8.

I've seen this with the young boys that used to play against the
computers in the old days at the Ing competition. These boys were like
5 or 6-dan amateur level players with private teachers. Certainly no
professionals yet, although obviously on their way. These kids never
seemed to even contemplate bad moves, or bad-style moves. They never
acquired the bad habits that players pick up at a later age. But their
tactical and strategic skills were only so-so. Apparently that gets
learned at a later age.

So rather than the perceptual level changing when moving to
professional level, what happens I believe is this: at very young age
the brain is particularly capable of learning certain types of
patterns. But not very good at abstract thinking, like tactical and
strategic thinking. When people learn to play at a later age, they
compensate their lack of pattern learning by tactical and strategic
thinking. Which actually gets in the way of the pattern learning. For
example, you shouldn't play an empty triangle unless in special
situations where it can give a big tactical gain. If your brain first
gets wired not to even consider the empty triangle, it can be
overriden by tactical thinking at a later age. If you start learning
the game through tactical thinking, including considering the empty
triangle, it wastes too much time and you have to break the habit of
considering it all the time. And breaking such habits is much harder
than not learning them in the first place.

    Mark



On Fri, Aug 3, 2012 at 8:31 AM, Peter Drake <drake at lclark.edu> wrote:
> An interesting line from this article on pattern matching, including
> computer Go:
>
> “The great cognitive scientist Herbert Simon, who won the Nobel Prize
> for Economics, recognised this as needing to build up chunks of little
> patterns, and needing at least 50 000 of these to reach expert level
> at anything. We now think it is more than 100 000 patterns.”
>
> http://news.csu.edu.au//director/latestnews.cfm?itemID=C712563AEEA8398B7655D520C442F510&printtemplate=release
>
> --
> Peter Drake
> https://sites.google.com/a/lclark.edu/drake/
> _______________________________________________
> Computer-go mailing list
> Computer-go at dvandva.org
> http://dvandva.org/cgi-bin/mailman/listinfo/computer-go



More information about the Computer-go mailing list