Hacker News Reader

Comment by nicpottier | Hacker News Reader

Fun read, but man, the hubris! I play chess a fair bit and have gone through various stages of study, though admittedly never a month full time. But man, no way do I think Max could beat even an intermediate player (say 1800 ELO) 3 out of 5 games. It just isn't something you pick up in a month.

View on HN

Some people are really good at chess. Some people are really good at self-promotion stunts.

As they say, it takes all kinds.

At least he developed his pieces. What I found interesting is that he decided to go the weird ML way, instead of spending the whole month solving tactics puzzles. The only thing he did that I recognize as advice that I already thought was good is playing against moderately stronger opponents and reviewing those games. As far as I know, chess and go take about a year of decent training to get to a "competent amateur" level.

That part really irks me as well. It sounds like a move for the cool factor that will go into a Medium post or book later, rather than a move to optimize for the best chance of beating Magnus.

Building a ML algorithm to beat Magnus is really not something to do in a month, not to mention replicating the ML algorithm by computing in his head.

I mean trying a nonconventional means of improving was the only thing in the article that made some sort of sense, since traditional methods of chess training would have certainly given him zero hope. Not that the ML method sounds remotely possible, but at least it gave the reader the ability to suspend some disbelief.

Reply on news.ycombinator.com