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

81–90 of 300 posts

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

#81

Post by bee_rider »

jgalt212 wrote: post #40
bee_rider wrote: post #9

Earlier quoted context omitted.

I have heard this about various games (Risk and Settlers of Catan mostly). I don’t get it. The point of these games is to be a bastard. If everyone goes into it with the understanding that there will be double crosses, how could there be hurt feelings? Monopoly is worse; it is just boring, I would dump my friends if they suggested monopoly not because I was hurt by their ruthless gameplay but for their terminally tas…

> The point of these games is to be a bastard. I guess you never play poker because you don't like to bluff. Or you've never faked a pass in basketball before taking a shot. Or you and the catcher like to announce when you're going to throw a curveball.

I’m not sure why you’d guess that, stringing somebody along for hours in Risk and then backstabbing them at the crucial moment seems to at least match your suggestions in terms of duplicity, and is the height of glorious bastarddom.

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

#82

Post by O__________O »

ekdinan wrote: post #62
O__________O wrote: post #12

Possible I missed it, but I don’t see any reference to how they addresses chat thread continuity (aka memory) issues that are major issue for AI chat bots. Did they address this and if so, how?

Indeed, consistency (with the lengthy dialogue histories, but also the game state as well as game plans) was a huge challenge for us. We spent a lot of time working on techniques for detecting and filtering these kinds of low quality messages. You can see the Supplementary Materials in the paper for full details, but TL;DR: we built a suite of classifiers for detecting common mistakes (training classifiers to discrim…

Congrats on getting the related research published.

Feel like a hack would have been to try to force dialogue into an extractable form that stored a state model relevant to the game, even additional hacks like asking the opposing player to restate their understanding of prior agreements; disclosure that I have no idea how the game Diplomacy works, so might be irrelevant.

Beyond that, no idea how Facebook manages its AI research, but quick Google confirms my memory that Meta/Facebook has done prior research on enabling AI memory capabilities related to recall, forgetting, etc.; which I mention just in case you were not aware.

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

#83

Post by euroderf »

mckeed wrote: post #75
euroderf wrote: post #73

I assume that in-house they sic'd several instances on each other. And I suspect that the result so horrified them that they dared not publish it in conjunction with the results under discussion here.

That's an interesting point, though – that compared to Chess or Go, playing against other versions of itself is less helpful for improving against humans, due to the requirement of communication with the other players.

Sure thing. But if they trained these against human players, winningly, and THEN sic'd'em on each other... oh dear.

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

#84

Post by colonCapitalDee »

MichaelCollins wrote: post #41
bee_rider wrote: post #9

Earlier quoted context omitted.

I have heard this about various games (Risk and Settlers of Catan mostly). I don’t get it. The point of these games is to be a bastard. If everyone goes into it with the understanding that there will be double crosses, how could there be hurt feelings? Monopoly is worse; it is just boring, I would dump my friends if they suggested monopoly not because I was hurt by their ruthless gameplay but for their terminally tas…

IMHO, Risk is one of the few board games worth playing specifically because it's so engaging. Most of my friends who are "into boardgames" tell me that Risk is a terrible game because it starts fights. But the games they want to play bore me to tears. (I can't even read the manual for Settlers of Catan without dying of boredom, but from what you say maybe I should give it another try.)

I can't stand Risk because Risk rewards inactivity. It is almost never optimal in Risk to attack another player (although limited skirmishes are fine), because even a successful assault will badly damage your own forces leaving you vulnerable. The way to win Risk is to convince other players to attack each other and then clean up in the aftermath. This is ok for players that are new to the game because they'll actually attack each other (they don't know any better). But experienced players will almost never attack, resulting in a stalemate.

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

#85

Post by gcanyon »

kneebonian wrote: post #5

For those hearing of this game for the first time and think it sounds fun. Be warned, it is considered to be a game that will end friendships that have lasted for years, and monopoly is considered to be a game that causes less bad feelings. An AI may be able to speak like a person but will never be able to hang onto that long burning simmering hatred from when Brad didn't support my army and instead flipped on me by…

The world champion seems to say that the best Diplomacy involves honesty and cooperation. Is he eliding that point toward the end where you use that honesty and cooperation to flip and really screw someone? Or is there more to Diplomacy than just "convince people you're cooperating until the very best time to betray them"?

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

#86

Post by vkou »

bee_rider wrote: post #9
kneebonian wrote: post #5

For those hearing of this game for the first time and think it sounds fun. Be warned, it is considered to be a game that will end friendships that have lasted for years, and monopoly is considered to be a game that causes less bad feelings. An AI may be able to speak like a person but will never be able to hang onto that long burning simmering hatred from when Brad didn't support my army and instead flipped on me by…

I have heard this about various games (Risk and Settlers of Catan mostly). I don’t get it. The point of these games is to be a bastard. If everyone goes into it with the understanding that there will be double crosses, how could there be hurt feelings? Monopoly is worse; it is just boring, I would dump my friends if they suggested monopoly not because I was hurt by their ruthless gameplay but for their terminally tas…

Settlers of Catan is in many respects not a very good game, because many of its mechanics are zero or negative-sum, with a heavy helping of RNG.

This creates a lot of opportunities for kingmaking and spiteful plays from people who are not able to win, but are able to make sure that you lose. And the worst part is that you're often forced into these situations through no fault of your own.

Diplomacy allows for all of the same plays, but as the player, you have way more agency about both getting into, and getting out of them. It makes the adversarial alliance-and-dealmaking part first and foremost.

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

#87

Post by gcanyon »

It seems like with this, Meta is doubling down on the Facebook platform's use as a dark influencer in politics. It's bad for humanity to enable advanced profiling and general targeting of demographics, but it's a whole other level if the communication itself can be targeted at the individual level.

We're headed for a reality where I don't know if my friends' friends are real or not, and they each entice me with arguments tailor-made to my sensibility to change my mind in ways that serve someone else's purposes.

Not to sound alarmist, but this is beyond Manhattan Project-levels of "maybe this won't turn out well for us."

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

#89

Post by teaearlgraycold »

mabbo wrote: post #28

I would love to see this kind of thing applied to an RPG. Randomly generate a city full of people. Make a few dozen of them the important NPCs. Give them situations and goals, problems they need to solve and potential ways to solve them. Certain NPC's goals are opposite others'. Then drop the player into that world and have the 'quests' the player is performing be generated based on the NPCs needing their help. Updat…

This is what a next generation Skyrim should be.

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

#90

Post by renewiltord »

ndonnellan wrote: post #14
kneebonian wrote: post #5

For those hearing of this game for the first time and think it sounds fun. Be warned, it is considered to be a game that will end friendships that have lasted for years, and monopoly is considered to be a game that causes less bad feelings. An AI may be able to speak like a person but will never be able to hang onto that long burning simmering hatred from when Brad didn't support my army and instead flipped on me by…

If you want all these things and exponential growth, check out Neptune's pride: https://np.ironhelmet.com/ for basically "diplomacy in space with a few extra dimensions". I haven't played in a few years (it was great at the beginning of the pandemic), but it was a lot of "fun" in the same way that diplomacy is "fun". Except with a space theme and a lot of anxiety waking up at 2am hoping your fleet arrives at an empty…

The last time we played this at the company, we lost maybe two weeks of productivity over the game period. I needed to run polyphasic sleep with 2 h naps throughout but my secret alliance got to the end game strongest and then it was a vicious little fight, and one of us took it.
Post reply on HN