[computer-go] Re: Optimizing UCT

2007-10-26 Thread Jason House
I know that I've missed at least two enhancements: 1. UCB1-Tuned (adding an upper confidence bound on the variance used in calculating upper confidence bounds) 2. First play urgency - Giving an artificial upper confidence bounds to untried moves (I've seen references that 110% win rate is the

Re: [computer-go] Re: Optimizing UCT

2007-10-26 Thread Jason House
Technical Report on MoGo link from http://www.lri.fr/~gelly/MoGo.htmeventually leads to http://hal.inria.fr/docs/00/12/15/16/PDF/RR-6062.pdf First-play urgency is described in the bottom paragraph on page 16. Results are shown 4 pages later in table 7 (influence of FPU). On 10/26/07, Magnus

Re: [computer-go] Re: Optimizing UCT

2007-10-26 Thread Magnus Persson
Quoting Jason House [EMAIL PROTECTED]: 2. First play urgency - Giving an artificial upper confidence bounds to untried moves (I've seen references that 110% win rate is the right setting) Where did you see that? -- Magnus Persson Berlin, Germany

Re: [computer-go] Re: Optimizing UCT

2007-10-26 Thread Olivier Teytaud
Technical Report on MoGo link from http://www.lri.fr/~gelly/MoGo.htm http://www.lri.fr/%7Egelly/MoGo.htm eventually leads to http://hal.inria.fr/docs/00/12/15/16/PDF/RR-6062.pdf http://hal.inria.fr/docs/00/12/15/16/PDF/RR-6062.pdf First-play urgency is described in the bottom paragraph on

Re: [computer-go] Re: Optimizing UCT

2007-10-26 Thread Jason House
On 10/26/07, Olivier Teytaud [EMAIL PROTECTED] wrote: MoGo is very different now. There's no UCT anymore in MoGo It's almost mean to tell us that MoGo isn't using UCT and omit further detail ;) Would it be safe to assume that it's UCB1-Tuned with the modifications discussed in the ICML