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

CICERO: An AI agent that negotiates, persuades, and cooperates with people

ai.facebook.com

Post reply on HN
Print view View original

291–300 of 300 posts

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#291

Post by mach1ne »

WastingMyTime89 wrote: post #290
mach1ne wrote: post #287

Earlier quoted context omitted.

What engineering development are you referring to? Do you have any examples?

Engines, planes, PVs. They all gradually became better in a way which doesn’t really map to the organic function they augment, replace or take inspiration from. Planes don’t fly like birds. There is very little reason for what you would call a “general intelligence” ai to develop in a way which mimics our own intelligence. It might but I would find that more notable that if it did not.

I think the biggest difference between building a general intelligence and other engineering problems is that in the latter, the target function is well-defined. With planes, anything that is able to fly safely goes. Nuclear reactors, anything that generates electricity. But with intelligence... What does it need to do, exactly? The 'moving goalpost' is moving because we have no idea where it should be. We're building AI blindfolded. I don't think we're currently solving the problem, simply because we don't know what the problem is.

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#292

Post by naasking »

mach1ne wrote: post #286
naasking wrote: post #277

Earlier quoted context omitted.

> We're developing ever more sophisticated narrow AI, not even taking baby steps to proper true AI. I'm fairly certain you cannot rigourously define "proper true AI", and I'm also fairly certain you don't mechanistically understand how humans think. This raises the question: where does your confidence that our current path is not pretty close to "proper true AI" and/or how humans think come from?

The only intelligence with human capabilities is, so far, human intelligence. We can see how nature incrementally improved 'the algorithm' when we look at other animals, from simpler ones to primates. We don't see the capabilities of even lower animals in modern deep learning based networks. We know these networks don't match in neural function what we see in biology. Whether deep learning will lead to the North Star…

> We don't see the capabilities of even lower animals in modern deep learning based networks

Machine learning models have written multiple scholarly papers that have been accepted to prestigious journals. Find me a rabbit that did that.

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#293

Post by bigwavedave »

sanxiyn wrote: post #252
nopinsight wrote: post #235

Earlier quoted context omitted.

"Having read the paper & supplementary materials, watched narrated game & spoken to one of the human players I'm pretty concerned. The @ScienceMagazine paper centres 'human-AI cooperation' & the bot is not supposed to lie. However, videos clearly show deception/manipulation" "Screenshots of the stab below. The human player said: "The bot is supposed to never lie [...] I doubt this was the case here" "I was definitely…

As far as I can tell, as described in the paper, the bot in fact never lies, in this sense: there is a model that generates messages from moves, where messages should correspond to moves, and when the bot says any messages, at the time, they are generated from moves the bot truthfully intends to play. On the other hand, the bot has no concept whatsoever of keeping its words. After saying words, it is free to change i…

> [snip] when the bot says any messages, at the time, they are generated from moves the bot truthfully intends to play.

> On the other hand, the bot has no concept whatsoever of keeping its words. After saying words, it is free to change its mind [snip]

Reminds of that one Asimov story about the robot who had a different interpretation of the first law of robotics. If my very hazy memory is right, the idea was that the robot could put a person in danger if it knew that it had the ability to prevent any damage from happening, but once it caused the danger, it could choose not to act and allow the person to come to harm.

I might be remembering this incorrectly, it's been a very long time since I read the story, but that was the first thought that came to mind when reading your comment :).

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#294

Post by andreyk »

sanxiyn wrote: post #257
andreyk wrote: post #176

This is super cool, but it's worth pointing that it still relies on supervised learning from a large dataset of human players, and that it's not a general learning algorithm (there is a whole bunch of problem-specific aspects to the model). It not having learning via RL (trial and error, as humans do) or self-play is kind of disappointing. Don't get me wrong, it's an impressive advance, but just as with AlphaGo it's…

Eh, it does learn from self play via RL. One section of the paper is literally titled "Self-play reinforcement learning for improved value estimation". Yes, that's only a small part of the entire system.

ah, good catch; I was going off of the blog post

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#295

Post by ImHereToVote »

naasking wrote: post #292
mach1ne wrote: post #286

Earlier quoted context omitted.

The only intelligence with human capabilities is, so far, human intelligence. We can see how nature incrementally improved 'the algorithm' when we look at other animals, from simpler ones to primates. We don't see the capabilities of even lower animals in modern deep learning based networks. We know these networks don't match in neural function what we see in biology. Whether deep learning will lead to the North Star…

> We don't see the capabilities of even lower animals in modern deep learning based networks Machine learning models have written multiple scholarly papers that have been accepted to prestigious journals. Find me a rabbit that did that.

But, but, but that is just tricks unlike human intelligence which is based on no trickery soul fuel.

/s

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#296

Post by ncr100 »

bo1024 wrote: post #102
pesenti wrote: post #20

Paper: https://www.science.org/doi/10.1126/science.ade9097 Code: https://github.com/facebookresearch/diplomacy_cicero Site: https://ai.facebook.com/research/cicero/ Expert player vs. Cicero AI: https://www.youtube.com/watch?v=u5192bvUS7k RFP: https://ai.facebook.com/research/request-for-proposal/toward... The most interesting anecdote I heard from the team: "during the tournament dozens of human players never even su…

Hmm, I guess facebook doesn't have to go through IRB for human subject experiments, nor does Science require it, apparently.

I didn't read the paper. Creating an AI to lie seems like the Wrong Path.

Zuckerberg should shut this down ASAP. If that's the case.

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#297

Post by jml7c5 »

taneq wrote: post #199
jml7c5 wrote: post #186

Earlier quoted context omitted.

"All is fair in Love and Civ" is a common variant of the default rules. It introduces the "your box has been pwned" victory/loss condition.

Is this “all’s fair” stance an actual accepted thing? If so, what are the limits on it, can I assassinate someone? Start an actual war to cause a player to resign? Blow up their planet? Clearly there’s a tacit assumption of some boundary around the game.

I hope it isn't a real thing! I was being facetious.

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#298

Post by mattnewton »

CallMeJim wrote: post #229
mattnewton wrote: post #109

Earlier quoted context omitted.

Diplomacy is quite a bit different because you 100% win by being the nicest asshole. Past a small mechanical threshold, there really is no way to make it up in strategy or luck, it is only the interpersonal relationships and convincing that wins games. That, coupled with the fact that the game takes a long time and encourages a lot of communication, means the recipe to get invested and blur the boundary between game…

What!? Santa Clause isn't real?

Whoops haha. It is the best Tim Allen movie though.

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#299

Post by Borrible »

californiadreem wrote: post #60

You know, I feel privileged (in an epistemologically doubtful way) that I once lived in a world where I could have confidence that I wasn't being fundamentally deceived by literal dei ex machina. I've understood for the last decade that this faith was on borrowed time as we bridged closer to overt technophrenia, but it's harrowing to be a witness to it. I can't help but wonder at what point I will inhabit a world ind…

>I will inhabit a world indistinguishable from that of a paranoid schizophrenic.

My wife has schizophrenia, which is well under control with medication. But about every year or two, when she wakes me up in the night to tell me she's panicking because someone hacked her smartphone and laptop etc., I know we have to adjust her medication for a few weeks. It was scary at first, but now we know the drill: a night or two without much sleep, no big problem. Of course I check smartphone, laptop etc. You never can be sure, can't you? Especially not if you've already been in trouble with credit card fraud twice.

I once asked her psychiatrist what he would say to his patients who believe they are being monitored in these times of Snowden and Co. He said it didn't make his job easier, but he would calm them down and adjust the medication. After all, he knew that his patients were sick.

So what do you think they or their owners/masters will do when All is Watched Over by Machines of Loving Grace in our/their future/reality?

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#300

Post by ImHereToVote »

mach1ne wrote: post #291
WastingMyTime89 wrote: post #290

Earlier quoted context omitted.

Engines, planes, PVs. They all gradually became better in a way which doesn’t really map to the organic function they augment, replace or take inspiration from. Planes don’t fly like birds. There is very little reason for what you would call a “general intelligence” ai to develop in a way which mimics our own intelligence. It might but I would find that more notable that if it did not.

I think the biggest difference between building a general intelligence and other engineering problems is that in the latter, the target function is well-defined. With planes, anything that is able to fly safely goes. Nuclear reactors, anything that generates electricity. But with intelligence... What does it need to do, exactly? The 'moving goalpost' is moving because we have no idea where it should be. We're buildin…

It's just a pattern matching glorified search engine! He screamed as his legs were converted into paperclips.
Post reply on HN