Could a computer be good at playing ASL?

[Honza]

Computers are good at playing chess because chess is principally a game of calculation. If a computer was powerful enough to play ASL do you think it could be good at it? Is ASL also a game of calculation or does some sort of artistic or intuitive skill come into it?

 

[zgrose]

Could a computer be good at playing ASL?

Yes, but not in 2024 and probably not in 2025. 2030? Not sure I'd bet against it....

 

[clubby]

Computers sent a rover to Mars I'm assuming with enough programming you could teach it asl.

Can you see the computer quitting on turn 3 because it rolled 3 straight box cars.

 

[fenyan]

ASL has an infinitely greater number of possible moves than chess, so I'd say not anytime soon.

 

[Psycho]

Computers sent a rover to Mars I'm assuming with enough programming you could teach it asl.

Can you see the computer quitting on turn 3 because it rolled 3 straight box cars.

Or throw the dice across the room

 

[zgrose]

ASL has an infinitely greater number of possible moves than chess, so I'd say not anytime soon.

Like Go...?

 

[zgrose]

Following up from my Facebook reply here...
When this prize is claimed, ASL should fall soon thereafter...

ARC Prize

ARC Prize is a $1,000,000+ nonprofit, public competition to beat and open source a solution to the ARC-AGI benchmark.

arcprize.org

 

[Honza]

Like Go...?

Computers can play Go now well enough to beat the masters.

 

[PresterJohn]

It would be interesting to set up machine learning for an AI to play ASL, but first the rules would have to be clearly defined.

 

[hongkongwargamer]

Not until the computer can deliver thoughts and comments about family, life, investments and politics with some originality.

I never sat down at VASL to just play ASL. Sometimes the chats are so good that we didn’t get to play ASL at all!

Connections are formed by sharing common experiences. ASL excels by delivering rich shared experiences. VASL facilitates such sharing on a laid back, hassle free and low cost platform.

My buddies are an essential part of my ASL experience.

 

[Steed]

I hope not.

 

[Lorenzo26]

is not Chat GPT playing ASL ? It's AI or not ?

 

[Lorenzo26]

by the way, try to immagine conversations about an ASL game with chatGPT watching the game ! You have 4 hours ;-)

 

[Fort]

Like Go...?

Go has more possible moves than Chess...but it's not even close to the same as ASL. Then, when you throw in the randomness, I don't think a computer in any of our lifetimes will be able to play at the level of the best humans.

 

[Old Noob]

"Roll your 1MC, Hal"
"I'm sorry, Dave. But I can't do that."

 

[zgrose]

Then, when you throw in the randomness, I don't think a computer in any of our lifetimes will be able to play at the level of the best humans.

Well, I don't know how old you are, but AGI seems doable in our lifetimes.
But maybe I'm a glass-half-full kind of guy.

Pretty sure the randomness actually makes it easier for the computer than the human, btw.

 

[PresterJohn]

Go has more possible moves than Chess...but it's not even close to the same as ASL. Then, when you throw in the randomness, I don't think a computer in any of our lifetimes will be able to play at the level of the best humans.

The randomness and calculating the odds "without emotion" of multiple options is what computer AI is very good at. The problem is codifying the rules so that they are exactly understood without having to ask three other AI if the move is legal and then sending an email to MMP.

 

[daveramsey]

Codifying Andrea Fantozzi's Play By Self (https://www.facebook.com/groups/358520268920548/permalink/1078447893594445) probably would be the best chance at getting something that would play a reasonable game. It would be a big task, but something like this I think is better suited than the smaller decision space of chess/go (which are of course huge in of themselves) - but Chess/Go's ease of ascertaining the game-score at any given point is many orders of magnitudes easier than ASL's so for me you'd need more of a programmed decision tree rather than a neural-net style approach.

 

[jtsjc1]

I would think the computer could present somewhat of a challenge but it can't reason like a human. It can't predict what the other players moves are the same as a human. It will use probabilities but for those unorthodox plays it won't be able to change its strategy with the speed of a human to counter a surprise move.

 

[zgrose]

...but it can't reason like a human

Yet. That's why the ARC-AGI prize is there. These LLMs that are all the rage right now aren't going to master ASL. The next "tier" of AI should be able to tackle the decision trees.