[Computer-go] Feature wish for LeelaZero

Hello, in particular to GCP and other LeelaZero developers, currently the games of the European Pro tournament (in Jena) are transmitted on KGS. Some users give LZ percentages, but ... it seems that LZ does not give expected scores (and deviations) for the game end. (CrazyStone had such a feature

Re: [Computer-go] Accelerating Self-Play Learning in Go

> Blog post: > https://blog.janestreet.com/accelerating-self-play-learning-in-go/ > Paper: https://arxiv.org/abs/1902.10565 I read the paper, and really enjoyed it: lots of different ideas being tried. I was especially satisfied to see figure 12 and the big difference giving some go features made

Re: [Computer-go] Accelerating Self-Play Learning in Go

On Fri, Mar 8, 2019 at 8:19 AM Darren Cook wrote: > > Blog post: > > https://blog.janestreet.com/accelerating-self-play-learning-in-go/ > > Paper: https://arxiv.org/abs/1902.10565 > > I read the paper, and really enjoyed it: lots of different ideas being > tried. I was especially satisfied to see

Re: [Computer-go] Accelerating Self-Play Learning in Go

>> contrary to intuition built up from earlier-generation MCTS programs in Go, >> putting significant weight on score maximization rather than only >> win/loss seems to help. This narrative glosses over important nuances. Collectively we are trying to find the golden mean of cost efficiency...