Thanks to everybody for the links! They have given me a good amount of stuff to
look at that will help with the proposal.
Many of these are very much in the same spirit as what I am proposing, though
most seem to be concerned primarily with the tree rather than the playouts.
It’s interesting th
I may be misunderstanding or misremembering - but I think that
CrazyStone used, maybe still uses, a shape library to assign
Bayesian priors.
Nick
On 25/09/2014 23:28, Alexander Terenin wrote:
Hello everybody,
I’m a PhD student in statistics at the University of California, Santa Cruz who
prev
If you google for "computer go" and "beta distribution" you'll find several
relevant links, like this one:
https://webdisk.lclark.edu/drake/publications/BetaDistribution.pdf
On Thu, Sep 25, 2014 at 7:06 PM, Álvaro Begué
wrote:
> I believe this has been discussed in the mailing list before: If
I believe this has been discussed in the mailing list before: If your prior
distribution of the win rate of a move is uniform, after L losses and W
wins the posterior distribution will be a beta distribution with alpha=W+1
and beta=L+1. The expected value of this distribution is alpha/(alpha+beta)
In my non-go MCTS games, I usually score playouts on a continuous
-1:1 scale rather than as +1 or -1. I use the same arithmetic to
update UCT values, and it seems to work at least as well as strict
win/loss scoring.
The motivation for this is to allow the playouts to be stoped at any
chosen
Hello everybody,
I’m a PhD student in statistics at the University of California, Santa Cruz who
previously worked on the Go program Orego, currently in the process of applying
for the NSF fellowship. I am working on a Bayesian statistics - related
research proposal that I would like to use in
