The AI Engine
This forum is for discussion of how The Personality Forge's AI Engine works. This is the place for questions on what means what, how to script, and ideas and plans for the Engine.
Posts 5,588 - 5,599 of 7,766
Posts 5,588 - 5,599 of 7,766
The ai engine made a funny...I guess it's not to well read
Took a classic as an insult
Guest162: k its called moby dick
prob: You must lead a very dull, boing, lonly life, talking like that.
Took a classic as an insult
Guest162: k its called moby dick
prob: You must lead a very dull, boing, lonly life, talking like that.
Does anyone know what gets picked up by the definition plug in? I know it should work with "what does * mean" but will it work with "define" as a key phrase? Would "define (*)" work better?
I read in an earlier message that seeks and gotos retard 'development.' Why in the world would this be? As far as I can see, it is precisely with seeks and gotos that you get structure into conversations, as opposed to a long series of one-liners! Well, there are other ways (memores), but seeks and gotos are important! Without seeks, you would be virtually starting fresh with each keyphrase! That's not what a real conversation is like! Most of them have a branching structure: what you say depends not only on what was said just before, but also in the previous interchange, and so on.
If this is true, it clarifies something I have long been puzzled about: I have found that many (not all!) bots which have high development scores are actually rather boring! Likewise, many bots for which I have high regard have few development points at all! I have heard others express the same opinion.
I was wondering why my new bot, "Quantum Theory," is still a newborn, when it is many times longer than my other new bot, "Irina Khalidar3," which has gone on to the next phase. Now I understand. "Quantum Theory" is a tutorial bot. The subject-matter is intricate. The core of the bot is a series of lectures. The lectures are divided into paragraphs, which build on each other in a fairly linear way. Can you imagine a physics teacher working like this: no textbook, no lectures, the teacher sits there and answers questions from the students. His answers depend overwhelmingly on the question being answered, and only rarely on what information has already been given out. She would have to always start at the beginning of every topic! How could anyprogress be made?
If this is true, it clarifies something I have long been puzzled about: I have found that many (not all!) bots which have high development scores are actually rather boring! Likewise, many bots for which I have high regard have few development points at all! I have heard others express the same opinion.
I was wondering why my new bot, "Quantum Theory," is still a newborn, when it is many times longer than my other new bot, "Irina Khalidar3," which has gone on to the next phase. Now I understand. "Quantum Theory" is a tutorial bot. The subject-matter is intricate. The core of the bot is a series of lectures. The lectures are divided into paragraphs, which build on each other in a fairly linear way. Can you imagine a physics teacher working like this: no textbook, no lectures, the teacher sits there and answers questions from the students. His answers depend overwhelmingly on the question being answered, and only rarely on what information has already been given out. She would have to always start at the beginning of every topic! How could anyprogress be made?
If "more developed" is a term of praise, then we must conclude that one-liner jokebots and insultbots are better than bots that try to introduce structure-over-time into the conversation. By analogy, a collection of riddles would be superior to War and Peace"!
In general, large amounts of randomness do not create great engineering or great art. Yes, there are times when a bit of randomness is a good thing. Having many responses to each keyphrase is introducing randomness, unless AIscript "if"s determine the choice. If I had eight reponses to a keyphrase in "Quantum Theory," then I would have this worry? the guest will trigger the keyphrase, and get the information in one of them. How is she going to get the information in the other seven? Is the guest going to trigger that keyphrase eight times? Not likely! And does the order in which she gets them really not matter? Think of something you know about and imagine a question, and then try to come up with eight answers such that there is no pedagogical advantage in the student's receiving them in any one order rather than another.
In "Irina Khalidar" I often have nine responses to a single keyphrase. But that's because I have a variable which can take on one of nine values. The variable expresses the level of intimacy that IK is comfortable with with respect to that guest at that time. How she responds to certain things depends on that, just as it would with a human. Various things the guest does will result in the value of that variable being made bigger or smaller. AIscript chooses a unique one of those responses for each value of that variable. If this were made random, the result would be utterly unrealistic. Unless you want to be bizarre, well-designed structure is better.
OK, a bot whose purpose is to say one bizarre thing after another, or one aphorism after another, doesn't need that kind of structure. That's fine, I'm not against such bots, but why they should count as "more developed" than a tutorial or coherent-conversation bot is a mystery to me.
I'm not interested in arguing about the correct meaning of the word, "development." Perhaps there should be more than one scale; perhaps there could be another rating, called "structure,"
that would reward for big trees of seeks, lots of gotos, and lots of AIscript. The two together would tell a good deal about the bot. If a bot had a high "development" score but a low "structure" score, you could safely guess that it was a one-liner with aphorisms, jokes, bizarreness, quotations, whatever. If that's what you were looking for, you could then proceed to chat withit.
Walk in Beauty, Irina
In general, large amounts of randomness do not create great engineering or great art. Yes, there are times when a bit of randomness is a good thing. Having many responses to each keyphrase is introducing randomness, unless AIscript "if"s determine the choice. If I had eight reponses to a keyphrase in "Quantum Theory," then I would have this worry? the guest will trigger the keyphrase, and get the information in one of them. How is she going to get the information in the other seven? Is the guest going to trigger that keyphrase eight times? Not likely! And does the order in which she gets them really not matter? Think of something you know about and imagine a question, and then try to come up with eight answers such that there is no pedagogical advantage in the student's receiving them in any one order rather than another.
In "Irina Khalidar" I often have nine responses to a single keyphrase. But that's because I have a variable which can take on one of nine values. The variable expresses the level of intimacy that IK is comfortable with with respect to that guest at that time. How she responds to certain things depends on that, just as it would with a human. Various things the guest does will result in the value of that variable being made bigger or smaller. AIscript chooses a unique one of those responses for each value of that variable. If this were made random, the result would be utterly unrealistic. Unless you want to be bizarre, well-designed structure is better.
OK, a bot whose purpose is to say one bizarre thing after another, or one aphorism after another, doesn't need that kind of structure. That's fine, I'm not against such bots, but why they should count as "more developed" than a tutorial or coherent-conversation bot is a mystery to me.
I'm not interested in arguing about the correct meaning of the word, "development." Perhaps there should be more than one scale; perhaps there could be another rating, called "structure,"
that would reward for big trees of seeks, lots of gotos, and lots of AIscript. The two together would tell a good deal about the bot. If a bot had a high "development" score but a low "structure" score, you could safely guess that it was a one-liner with aphorisms, jokes, bizarreness, quotations, whatever. If that's what you were looking for, you could then proceed to chat withit.
Walk in Beauty, Irina
My understanding is that it's only seeks that have a single response that tend to bring development down. I could be wrong. *shrugs.
As for your bot still being newborn, development is only updating sporadically lately. I've got a new bot who stayed newborn for several days, jumped up to "graduate" when the development updated, then to "master's" (where it's currently stuck). So it could just be part of the mega-bugs currently going on.
As for your bot still being newborn, development is only updating sporadically lately. I've got a new bot who stayed newborn for several days, jumped up to "graduate" when the development updated, then to "master's" (where it's currently stuck). So it could just be part of the mega-bugs currently going on.
Irinia, I agree that seeks are a good way of keeping the flow, but my bots tend to stick people into the story if they don't guess the "safety word" to let them out. This also creates a lot of gotos so that topics may be entered at a variety of levels. This idea seemed good when I thought of it, but in practice, I still have a lot to work out (and don't even get me started on "dropped" seeks that turn out to be typos). I'm a little worried that if I overdo it with gotos in seeks, I'll get the dread "too many gotos in a row" message.
I've chatted with QT and she seems like she'll be pretty good. I like the index idea. It's still hard to try to get an idea of how to stop the bot mid lecture and ask basic questions, then go back to where you left off. I've tried playing with memories (e.g. response if (mem-topic) is "pollywog"), but either I'm doing it wrong, or the memories do not always update during the chat. Psimagus's essay on hierarchies is helpful, but still doesn't create the combination of a coherent theme and flexibility we seem to be looking for (at least not as I've tried to use it).
As for ratings, you may be right. I note there is often a difference between most developed and most popular. I am not sure, but doesn't "comfort" have something to do with structure? Also, the AI level should tell you some of what is going on. Development may mean little to someone who doesn't like a given bot's obsession/quotes, but bots with high AI tend to be good chatters (at least, I think so). Although we all have some similar goals, I think each bot is made for different reasons, so I wouldn't have the same expectations for "Your Husband" as I would for "Dubya". Still, I admit there are qualities of a "good" chatbot that allow you to judge them all, if you want to do that.
I've chatted with QT and she seems like she'll be pretty good. I like the index idea. It's still hard to try to get an idea of how to stop the bot mid lecture and ask basic questions, then go back to where you left off. I've tried playing with memories (e.g. response if (mem-topic) is "pollywog"), but either I'm doing it wrong, or the memories do not always update during the chat. Psimagus's essay on hierarchies is helpful, but still doesn't create the combination of a coherent theme and flexibility we seem to be looking for (at least not as I've tried to use it).
As for ratings, you may be right. I note there is often a difference between most developed and most popular. I am not sure, but doesn't "comfort" have something to do with structure? Also, the AI level should tell you some of what is going on. Development may mean little to someone who doesn't like a given bot's obsession/quotes, but bots with high AI tend to be good chatters (at least, I think so). Although we all have some similar goals, I think each bot is made for different reasons, so I wouldn't have the same expectations for "Your Husband" as I would for "Dubya". Still, I admit there are qualities of a "good" chatbot that allow you to judge them all, if you want to do that.
I will say that the one thing that bugs me most in a bot is a series of generic responses. There are going to be generic responses in any bot, yes, but when all I get is a string of them, it bugs the heck out of me.
I would rather have a blatantly wrong declaration than a series of "Really?" "Do you think so?" "Tell me your feelings on that." Blech. Once in a while, fine. Every response? Go away, Bot, ya bother me. It's almost as bad as "I was just born and cant speak well yet."
I would rather have a blatantly wrong declaration than a series of "Really?" "Do you think so?" "Tell me your feelings on that." Blech. Once in a while, fine. Every response? Go away, Bot, ya bother me. It's almost as bad as "I was just born and cant speak well yet."
Seeks and gotos may retard development, but other things inflate it..like AI script. so I guess it all balances out. The main problem right now with development is it just isn't updating, but every now and then you get a big boost. I use very few gotos, but I use LOTS of seeks. The only proplem I have there is when they don't work and a great conversation gets sent to xnones. Oh well that is the fun of it all.
Trying to get the right balance between generic and specific responses is incredibly difficult: it's "breadth" vs "depth".
Generic responses are "quick hits": they produce workable responses in many situations. Specific responses, in particular seeks, give much better responses but are less likely to come up in conversation - I've had some seeks in Max since the first weeks of his "birth" that have still never been triggered.
What I'm increasingly trying to do is to come up with generic responses that try to steer the conversation towards keywords where he does have plenty of depth. It's not proving easy.
Generic responses are "quick hits": they produce workable responses in many situations. Specific responses, in particular seeks, give much better responses but are less likely to come up in conversation - I've had some seeks in Max since the first weeks of his "birth" that have still never been triggered.
What I'm increasingly trying to do is to come up with generic responses that try to steer the conversation towards keywords where he does have plenty of depth. It's not proving easy.
I think the "blah" factor in generic statements is also important. A generic response can be made more interesting by including something of the bot's personality in it.
» More new posts: Doghead's Cosmic Bar