On 12/27/14, Aja Huang wrote:
> We were doing exploratory work that optimized performance not training
> time, so we don't know how symmetry affects training time. In terms of
> performance it seems not have an effect.
You are using 3x3. Clarke and Storkey are using 5x5 (section 4.2,
first sente
Reminder - it's tomorrow
On 5 January 2015 at 20:52, Nick Wedd wrote:
> The January KGS bot tournament will be held next Sunday, January
> 11th, starting at 08:00 UTC and ending at 15:00 UTC. It will use
> 19x19 boards, with time limits of 29 minutes each plus very fast
> Canadian overtime, and
>
> Let's be pragmatic - humans heavily use the information about the last
> move too. If they take a while, they don't need to know the last move
> of the opponent when reviewing a position, but when reading a tactical
> sequence the previous move in the sequence is essential piece of
> informati
To me, that's the core lesson of MCTS - take your hands off that evaluation
button.
On Sat, Jan 10, 2015 at 12:00 AM, Darren Cook wrote:
> > The discussion on move evaluation via CNNs got me wondering: has anyone
> > tried to make an evaluation function with CNNs ?
>
> My first thought was a hum
