[Computer-go] Maximum Frequency method
hideki_katoh at ybb.ne.jp
Sat Sep 5 16:57:40 PDT 2015
uurtamo .: <CADg0iNBRQg9d1MY2VWEJ_n4GkRZbL5iOYO6Bscrh7mPhFCsQJA at mail.gmail.com>:
>BTW: have you tried other distributional difference metrics, or does K-L
>have properties that you like?
No. #I haven't tried any distributional difference. The paper is not
As MC simulations add big randomness to the scores, I wonder if
maximizing frequency method is better than simple average. My
expectation is it's almost the same. Improving simulations has much
more priority for me.
#Mathematically, K-L divergence would be the best measure. anyway.
Problem is maximizing frequency may not approximate maximizing the
divergence due to the simulations' random biases.
>On Sep 5, 2015 1:35 AM, "Hideki Kato" <hideki_katoh at ybb.ne.jp> wrote:
>> djhbrown .: <
>> CAPSify9fub60pD3LZdyHDPupfFgYeNv4t+M47okWPHzrB4Qxgw at mail.gmail.com>:
>> >thank you for sharing the paper.
>> >"the Maximum Frequency method is based on the
>> >maximization of the difference between the expected reward of
>> >the optimal move and that of others"
>> >intuitively it feels that biasing random search towards the optimal route
>> >would yield reduced failure rates, yet it does seem to depend on knowing
>> >what the optimal route is beforehand.
>> UCT is never a random search but deterministic.
>> Maxmizing KL-divergence just speed-up the convergence of the interative
>> >if i knew the optimal route to get from A to B, i wouldn't bother doing a
>> >random search, but just follow it.
>> >"This property [bias in suboptimal moves] means that the impact of
>> >missing the optimal move is much greater for one player than it is for the
>> >i find this conclusion puzzling because Go is a zero-sum game, so what is
>> >good for one side is equally bad for the other, not variably so. I have
>> >not checked the statistical inference calculations to see whether there is
>> >an error in them.
>> >---- inline file
>> >Computer-go mailing list
>> >Computer-go at computer-go.org
>> Hideki Kato <mailto:hideki_katoh at ybb.ne.jp>
>> Computer-go mailing list
>> Computer-go at computer-go.org
>---- inline file
>Computer-go mailing list
>Computer-go at computer-go.org
Hideki Kato <mailto:hideki_katoh at ybb.ne.jp>
More information about the Computer-go