~marginalrevolution | Bookmarks (6)

The amazing Gukesh (17 years old!)  is half a point ahead with one round remaining today.  Three players — Nakamura, Caruana, and Nepo — trail him by half a point.  Naka is playing Gukesh, and of course Naka will try to win.  But what about Caruana vs. Nepo?  Yes, each must try to win (no […] The post The game theory of the final...

For the first time in 75 years, Mattel is making a major change to the iconic board game Scrabble — and touting a “No More Scoring” gameplay option. The new launch is a double-sided version of the famous board game — one side with the original game for those who want to stick to the […] The post Scrabble markets in everything, okie-dokie edition...

  I agree Caruana is the clear favorite, but at these odds he seems slightly overvalued and Nepo undervalued?  Keep in mind the margins of quality here are slight, and furthermore Nepo won it the last two times.  The tournament could in fact come down to whoever can beat Nijat twice.  While that correlates with […] The post The Candidates tournament appeared first on...

Remember when man and machine played together to beat the solo computers?  It was not usually about adding the man’s chess judgment to that of the machine, rather the man would decide which computer program to use in a given position, when the programs offered conflicting advice. that was called Centaur Chess, or sometimes “Freestyle […] The post “Centaur chess” is now run by...

As you may know, a major tournament is going on right now, based on a variant of Fischer Random rules, sometimes misleadingly called “Freestyle.”  Subject to some constraints, the pieces are placed into the starting position randomly, so in Fischer Random chess opening preparation is useless.  You have to start thinking from move one.  This […] The post Why I don’t like Fischer Random...

Scaling seems to be working here: The recent breakthrough successes in machine learning are mainly attributed to scale: namely large-scale attention-based architectures and datasets of unprecedented scale. This paper investigates the impact of training at scale for chess. Unlike traditional chess engines that rely on complex heuristics, explicit search, or a combination of both, we […] The post Grandmaster-Level Chess Without Search appeared first...