Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Adversarial Wordle

qntm.org

Post reply on HN
Print view View original

61–70 of 72 posts

Re: Adversarial Wordle

#61

Post by MereInterest »

salomon812 wrote: post #50
vitus wrote: post #44

Earlier quoted context omitted.

It's worth noting that there are two dictionaries used in this implementation: "possibleWords" (from which the solution can be drawn) and "impossibleWords" (other words that are valid guesses). https://qntm.org/files/wordle/words.js These are most likely the same words in the implementation at https://www.powerlanguage.co.uk/wordle/ based on ["cigar","rebut","sissy","humph","awake","blush","focal","evade", showing up…

Good point! And it's clear to me that this isn't the dictionary I was using. I grabbed a random Scrabble dictionary online and filtered for length of five. I'll rework my code to take all this into account and we'll see what happens (it probably will happen later today.)

I'm guessing that we're both working from the 2019 Collins Scrabble Word list[0], since that's the list of words I used, my implementation also tries to minimize the maximum number of resulting possibilities, and also have SERAI as the first guess every time.

[0] https://boardgames.stackexchange.com/questions/38366/latest-...

Re: Adversarial Wordle

#62

Post by salomon812 »

christiangenco wrote: post #56

My solver at wordlesolver.com plays the same 5-move game each time: ARISE, BLUDY, COMET, CHUNK, CHUCK. It looks like optimal play is 4 moves: https://twitter.com/zwegner/status/1480037275803308034

Your solver is awesome!

I'm really curious what the algorithm is find the 4 move play. Mine only optimizes for a single look-ahead.

Re: Adversarial Wordle

#63

Post by salomon812 »

zwegner wrote: post #57

I posted this on Twitter last night, but here's my (purported) proof that 4 guesses is optimal: https://gist.github.com/zwegner/508cc183ab94dd27686a40384783... ...and a bit more explanation in the Twitter thread: https://mobile.twitter.com/zwegner/status/148011092775217561...

Ah, apologizes, I missed your post. This analysis is awesome! I did a single look ahead, so it falls into the -OUCH zone and I have to burn some guesses. Doing the full search across all guesses is key. Well done!

Re: Adversarial Wordle

#64

Post by re »

MattRix wrote: post #45
raviparikh wrote: post #32

I made my own adversarial wordle just now, not realizing this one already existed: https://swag.github.io/evil-wordle

they are both cool but right now they both have the problem that they don’t recreate the wordle keyboard, which is a big part of what makes the game enjoyable to play.

Looks like qntm just updated the UI to include a Wordle-style keyboard (and also renamed it from "Adversarial Wordle" to "Absurdle") in the last few minutes.

Re: Adversarial Wordle

#65

Post by MerelyMortal »

brk wrote: post #34
indigodaddy wrote: post #27

So I don’t really get this exactly.. I guessed twice and then it solves it for me??

It sounds like you hit the "Give Up" button by accident.

I accidently tapped the "Give Up" button, and it says I guessed correctly. Unless the correct move is not to play, there seems to be a bug.

Re: Adversarial Wordle

#66

Post by whaatt »

MattRix wrote: post #45
raviparikh wrote: post #32

I made my own adversarial wordle just now, not realizing this one already existed: https://swag.github.io/evil-wordle

they are both cool but right now they both have the problem that they don’t recreate the wordle keyboard, which is a big part of what makes the game enjoyable to play.

Self-promoting my own take on this idea (https://skalon.com/2022/01/08), which basically implements it within the familiar Wordle interface.

Was quite surprised to see two other takes on this idea when I was getting ready to post this! IIRC there are a few different undergrad CS programs that have you implement something like this for hangman.

Post reply on HN