I think this is a good step forward that generalizes an algorithm to play both perfect and imperfect information games. However, table 9 shows (I believe it shows, it is not the most intuitive form), that other AIs (Deepstack, ReBeL, and Supremus) eat its lunch at poker. It also performs worse than AlphaZero at perfect information games. So, while a nice generalizing framework, probably will not be what you use in practice.
Player of Games
141–150 of 242 posts
Re: Player of Games
#142Re: Player of Games
#143Comparing against Stockfish 8 in a paper released today and labeling it as "Stockfish" is bordering on being dishonest. The current stockfish version (14) would make AlphaZero look bad, so they don't include it ...
The first mention says "Stockfish 8, level 20" in the paper. This isn't a blog post that you can skim, you need to read the whole thing before critiquing.
Re: Player of Games
#144Earlier quoted context omitted.
The name of the game here is generality. For a really general agent, they are looking to have superhuman performance, not get state of the art on every individual task. Beating stockfish 8 convinces me that it would be superhuman at chess.
They could still be honest that it's Stockfish 8, not the Stockfish everyone has. Your product having genuine value does not excuse lying about that value.
I observed this kind of behavior in many papers nowadays. This extremely painful for research, because some better candidates could be overseen and FAANG publishs a majority in the ML-paper section. Its a mess.
Re: Player of Games
#145Earlier quoted context omitted.
My recollection is that by the end of the novel its clear that Gurgeh was never competitive with the ship, although he might have been competitive with his security drone (although even that isn't clear, since imply that the security drone is a better game player than it pretends to be). To me it felt like the whole point of the novel was that Gurgeh was a piece in an even larger game and he didn't even realize it. S…
Yeah, I thought it was clear from the beginning of the book that no humans were even remotely competitive with any AI (including the main character) but that human game players were sort of an aesthetic throwback, like dog-racing in an era of F1 cars.
This was my understanding as well, but I might have read into it. The culture minds are in freaking hyperspace to get around lightspeed limitations on computations. He for sure can't beat that, but he could beat someone on another planet at their own game that he literally just learned in the year it took to get there. A game that permeates every aspect of their civilization.
I do assume his drone could beat him as well, but I'm not sure.
Re: Player of Games
#146Earlier quoted context omitted.
The name of the game here is generality. For a really general agent, they are looking to have superhuman performance, not get state of the art on every individual task. Beating stockfish 8 convinces me that it would be superhuman at chess.
They could still be honest that it's Stockfish 8, not the Stockfish everyone has. Your product having genuine value does not excuse lying about that value.
They were? They say they use Stockfish 8 the very first time they mention it.
Re: Player of Games
#147Earlier quoted context omitted.
the same goes for slumbot in poker, its super old like 2013, the game is played completely different now and current bots would destroy it.
As a commenter above noted, this work is about generality, being able to play every game, and not being the best at every game.
As noted before, the reason for including old tech is to look better. Why not mention the current state of the art and show that with a general player we can come close to this results?
This is just benchmark cherry picking and does not reflect real performance or comparison.
Re: Player of Games
#148Earlier quoted context omitted.
How did I miss this plot point? It's been a while, but I remember focusing on the game Gurgeh played. Maybe I just don't remember it now.
The last page of the book gives it away: (MASSIVE SPOILER OBV) The security drone who came with Gurgeh to Azad was the same drone he meets during the introduction chapters who was "rejected" from SC and offers to let him cheat (though it was wearing a disguise at the time). Then, after he cheats he basically gets blackmailed into going to Azad and conveniently this "non-SC" drone comes with him in a very "non-SC" shi…
Yes, at the end you start to question just who the player of games actually was.
Re: Player of Games
#149Comparing against Stockfish 8 in a paper released today and labeling it as "Stockfish" is bordering on being dishonest. The current stockfish version (14) would make AlphaZero look bad, so they don't include it ...
The first mention says "Stockfish 8, level 20" in the paper. This isn't a blog post that you can skim, you need to read the whole thing before critiquing.
That's actually the second mention, the first is when they introduce the games in section 4:
> Today, computer- playing programs remain consistently super-human, and one of the strongest and most widely-used programs is Stockfish.
They also go back to referring to it as Stockfish for the rest of the paper.
An analogous situation in my mind would be if AMD released a new CPU and benchmarked it against an Intel CPU, only mentioning once, somewhere in the middle of the paper, that it was a Pentium 4.
Re: Player of Games
#150Earlier quoted context omitted.
Why?
I don't know how go changed. But as for chess the tournament play at the master level have insane deep opening preparation done before with computers. They play preparation game where they try to guess what lines the opponent checked and memorized before the games. They aren't actually playing until their computer backed preparation ends more than the few moves that they have fed in to come up with something differen…
The change in Go is that professionals now play more AI-like stuff, basically the same opening moves, and that's fairly boring to watch. It's the best sequences of moves we have so far, but also everyone knows them and isn't interested.
Another change is that territory is more valued than influence now, which too makes games less fun to watch, at least in my experience. To my knowledge, Shibano Toramaru, professional Go player, used to play highly focused on influence, and his games were very interesting to watch; it was just spectacular. But after AlphaGo came he converted to focus on territory like everyone else, only occasionally letting his beast out. But I watched only a few videos on him so don't take my word; it's just my impression.