I want to implement a conversation system into my RPG (trying to get advanced AI as possible). Conversation as in, the player types:
"Hi, I would like a beer"
and the bartender would respond with
"Coming right up"
and then hand the player a beer.
I've got some ideas and some things I'd like to try, but first I would like to look at what's already been done. But extensive Googling does not turn up anything, so I'm wondering: has this been done or is there research being done in it? (I know this is very complicated, but I'm willing to give it a shot.)
Sure it has. Have a look at the "Eliza" program and its descendants. There's also a Wiki article on chatterbots that might interest you. Have a look at AIML as a way to represent the rules you might use.
For an advanced design, look up the game "Façade". The game's site describes the technologies used and gives links to relevant papers. There was also recently an extensive article in Gamasutra about this, called Beyond Façade: Pattern Matching for Natural Language Applications.
You may also want to look into the Turing Test and it's relevant scientific following/conferences/publications to see what has been done in the humanizing of AI speech.
Related
I'd like to write an application, which would imitate a player in an online game.
About the game: it is a strategy, where you can:
train your army (you have to have enough resources, then click on a unit, click train)
build buildings (mines, armories, houses,...)
attack enemies (select a unit, select an enemy, click attack)
transport resources between buildings
make researches (economics, military, technologic,...)
This is a simplified list and is just an example. Main thing is, that you have to do a lot of clicking, if you want to advance...
I allready have the 'navigational' part of the application (I used Watin library - http://watin.sourceforge.net/). That means, that I can use high level objects and manipulate them, for example:
Soldiers soldiers = Navigator.GetAllSoldiers();
soldiers.Move(someLocation);
Now I'd like to take the next step - write a kind of AI, which would simulate my gaming style. For this I have two ideas (and I don't like either of them):
login to the game and then follow a bunch of if statements in a loop (check if someone is attacking me, check if I can build something, check if I can attack somebody, loop)
design a kind of scripting language and write a compiler for it. This way I could write simple scripts and run them (Login(); CheckForAnAttack(); BuildSomething(); ...)
Any other ideas?
PS: some might take this as cheating and it probably is, but I look on this as a learning project and it will never be published or reselled.
A bunch of if statements is the best option if the strategy is not too complicated. However, this solution does not scale very well.
Making a scripting language (or, domain specific language as one would call that nowadays) does not buy you much. You are not going to have other people create AI agents are you? You can better use your programming language for that.
If the strategy gets more involved, you may want to look at Bayesian Belief Networks or Decision Graphs. These are good at looking for the best action in an uncertain environment in a structured and explicit way. If you google on these terms you'll find plenty of information and libraries to use.
Sounds like you want a finite state machine. I've used them to various degrees of success in coding bots. Depending on the game you're botting you could be better off coding an AI that learns, but it sounds like yours is simple enough not to need that complexity.
Don't make a new language, just make a library of functions you can call from your state machine.
Most strategy game AIs use a "hierarchical" approach, much in the same way you've already described: define relatively separate domains of action (i.e. deciding what to research is mostly independent from pathfinding), and then create an AI layer to handle just that domain. Then have a "top-level" AI layer that directs the intermediate layers to perform tasks.
How each of those intermediate layers work (and how your "general" layer works) can each determined separately. You might come up with something rather rigid and straightforward for the "What To Research" layer (based on your preferences), but you may need a more complicated approach for the "General" layer (which is likely directing and responding to inputs of the other layers).
Do you have the sourcecode behind the game? If not, it's going to be kind of hard tracing the positions of each CPU you're (your computer in your case) is battling against. You'll have to develop some sort of plugin that can do it because from the sound of it, you're dealing with some sort of RTS of some sort; That requires the evaluation of a lot of different position scenarios between a lot of different CPUs.
If you want to simulate your movements, you could trace your mouse using some WinAPI quite easily. You can also record your screen as you play (which probably won't help much, but might be of assistance if you're determined enough.).
To be brutally honest, what you're trying to do is damn near impossible for the type of game you're playing with. You didn't seem to think this through yet. Programming is a useful skill, but it's not magic.
Check out some stuff (if you can find any) on MIT Battlecode. It might be up your alley in terms of programming for this sort of thing.
First of all I must point out that this project(which only serves educational purposes), is too large for a single person to complete within a reasonable amount of time. But if you want the AI to imitate your personal playing style, another alternative that comes to mind are neural networks: You play the game a lot(really a lot) and record all moves you make and feed that data to such a network, and if all goes well, the AI should play roughly the same as you do. But I'm afraid this is just a third idea you won't like, because it would take a tremendeous amount of time to get it perfect.
I've been studying hierachial reinforcement learning problems, and while a lot of papers propose interesting ways for learning a policy, they all seem to assume they know in advance a graph structure describing the actions in the domain. For example, The MAXQ Method for Hierarchial Reinforcement Learning by Dietterich describes a complex graph of actions and sub-tasks for a simple Taxi domain, but not how this graph was discovered. How would you learn the hierarchy of this graph, and not just the policy?
In Dietterich's MAXQ, the graph is constructed manually. It's considered to be a task for the system designer, in the same way that coming up with a representation space and reward functions are.
Depending on what you're trying to achieve, you might want to automatically decompose the state space, learn relevant features, or transfer experience from simple tasks to more complex ones.
I'd suggest you just start reading papers that refer to the MAXQ one you linked to. Without knowing what exactly what you want to achieve, I can't be very prescriptive (and I'm not really on top of all the current RL research), but you might find relevant ideas in the work of Luo, Bell & McCollum or the papers by Madden & Howley.
This paper describes one approach that is a good starting point:
N. Mehta, S. Ray, P. Tadepalli, and T. Dietterich. Automatic Discovery and Transfer of MAXQ Hierarchies. In International Conference on Machine Learning, 2008.
http://web.engr.oregonstate.edu/~mehtane/papers/hi-mat.pdf
Say there is this agent out there moving about doing things. You don't know its internal goals (task graph). How do you infer its goals?
In way way, this is impossible. Just as it is impossible for me to know what goal you had mind when you put that box down: maybe you were tired, maybe you saw a killer bee, maybe you had to pee....
You are trying to model an agent's internal goal structure. In order to do that you need some sort of guidance as to what are the set of possible goals and how these are represented by actions. In the research literature this problem has been studied under the terms "plan recognition" and also with the use of POMDP (partially observable markov decision process), but both of these techniques assume you do know something about the other agent's goals.
If you don't know anything about its goals, all you can do is either infer one of the above models (This is what we humans do. I assume others have the same goals I do. I never think, "Oh, he dropped his laptop, he must be ready to lay an egg" cse, he's a human.) or model it as a black box: a simple state-to-actions function then add internal states as needed (hmmmm, someone must have written a paper on this, but I don't know who).
Anyone have good book / article recommendation for procedural generation of background music? (No vocals, just instruments).
I'm not interested in:
How do I generate the sound of a particular note on a particular instrument
I'm interested in:
How do I generate the melody / score for the music.
Thanks!
EDIT:
Thanks for the reference to Brian Eno. I'm definitely looking into the ambient/user can ignore type of music. I.e. think the background music of a game. It's there to provide some basic mood, but the focus is the game.
Sometime ago I ran into ChucK, which is a programming-language to generate music/sound/audio:
ChucK presents a new time-based, concurrent programming model that's highly precise and expressive (we call this strongly-timed), as well as dynamic control rates, and the ability to add and modify code on-the-fly. In addition, ChucK supports MIDI, OSC, HID device, and multi-channel audio. It's fun and easy to learn, and offers composers, researchers, and performers a powerful programming tool for building and experimenting with complex audio synthesis/analysis programs, and real-time interactive control.
I believe the end result can be converted into MIDI, which can then be converted into a score or sheet notation.
I don't know if this is what you're looking for. Hope this helps!
EDIT
After thinking about this a little longer, I think what you can possibly do (and this sounds a bit crazy) is write code that generates ChucK code. So define a set of rules for your music/score generation and then use that to create valid ChucK code. After you run the ChucK code, you can get a MIDI file which you can then convert into score/sheet-music.
The book "Computer Models of Musical Creativity" by David Cope should help you along with the theoretical side of computer-assisted composition, though you might want some music theory under your belt before you dive in.
If you are interested in procedural music check out the Condition30 site -- condition30.com
This music is all procedural.
If you're interested in an implementation of procedural music based on cellular automata in C#, you could grab the source code from http://proceduralmidi.codeplex.com/. A binary is also available.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 7 years ago.
Improve this question
I know that it's still an open problem so I don't expect to see complete answers here. I just want to find some approaches to solve the next problem:
I have a model (assume that is's bot's memory), and different words are associated with different objects in the model. Speaking with the bot is like executing sql-queries with a DB. Language is a very hard formalizable protocol. And we can't just write a million lines of code to implement some real language. But I believe that it's absolutely possible to implement some self-learning mechanism. How can it be implemented? Is it possible to implement learning "from scratch" or "from few basic words"? Just want to hear your ideas.
Actually, English is a very strict language and it's one of the easiest languages for experimenting with AI. Many other languages allow you to change the order of words (for example). And in some cases changed order can change the whole meaning or just add some intonation. I really don't have any ideas how to teach a bot for these things.
The first step, in taking this game to the next level, is ...
...to have a very clear view of prior art!
(and pardon me to say, the question doesn't suggest that you have such an extensive insight into the matter [and you're not alone, count me in ;-)])
Even, and maybe in particular, if your intention is to apply completely novel techniques and models, it seems important to review the literature on current and past practices. Aside from possibly identifying elements that may be adapted or reused in a new implementation, a survey of the domain will provide an keen understanding of the nature of the problem[s].
I've personally tried -on various and multiple occasions!- either the naive approach or the sophomoric approach to tackling broadly-defined problems. With the naive approach, one has but a very slight idea of the true nature and scope of the problem. The sophomoric sees us better equipped with domain knowledge and also with related tools, but this can also be misleading because without a deeper understanding, we tend to mis-read/mis-understand new material offered to us and also misuse some of the tools (a bit like the the fellow who's "good with a hammer" for whom many things look like a nail...)
It is particularly easy to make these mistakes in the field of NLP. That's because
Common sense seems to be all is required: after all a child, who's native tongue is English understands subtleties like
"He's not really an expert"
"He's really not an expert" (small wink at the OP's reference to the ordering of word in the English language)
We live in such exciting times, technology and knowledge wise: Processing power, programming language and tools, mathematical techniques, availability of affordable corpora... to name a few of these things that make this moment in time so special.
Far from me the idea of discouraging you in your chat-bot endeavor, I just hope that this long and generic exposé will encourage to look-before-you-leap, as this will truly save you time in the long run, I think in two ways:
provide you some frames of references (again, even if your intention is to "think outside these boxes")
maybe entice you to redefine the problem, for example by limiting it to particular domains of conversation (sports, or health, or life at a particular university campus...) or by focusing on a particular aspect of the problem (semantic awareness, smooth, natural sounding grammar, use of colloquial forms...)
Good luck ;-)
Check out MegaHAL's implementation for some ideas. We've used a variant of this bot for ages in an IRC channel of ours, and he does on occasion appear to be the intelligent mixture of many of our dominant personalities.
You "train" the bot -
each time the bot answer, you rank (or the tester) the answer - if the answer is good/logical - give high rank, if the answer is bad... low/negative rank.
use the ranking in the future to choose the answer, and this is how the bot learns...
There's a great description of Eliza in Paradigms of AI Programming. You should be able to implement a simple Eliza bot in a few days of work.
This isn't a learning algorithm, but it's surprising how realistic answers can be from something so simple.
You can create your own chat bot on BOT libre, http://www.botlibre.com.
The bots learns, can be trained, can be scripted, and your can program them, or let them program themselves.
Thew site supports embedding your bot on your own site, has REST API access, Android, IRC, Twitter. Free hosting, even for commercial bots.
AIML from the AliceBot project may help you out. It's a whole XML schema (if that doesn't put you off) for the branch of AI its concerned with.
An example from Wikipedia:
<category>
<pattern>WHAT IS YOUR NAME</pattern>
<template>My name is <bot name="name"/>.</template>
</category>
RebbeccaAIML is one quite well documented implementation.
I would like to know if there are any tools that can help me model C applications i.e. Functional programming.
E.g. I'm currently building a shared library.
But to communicate my design visually, I need something like UML. I would like to do this so that the person reviewing my design need not read through 100s of pages of functions, variables and so on.
I have read about UML for C, which I'm considering.
If there is anything better out there, please let me know.
The bottom line is to visualize the design of C applications and modules without reading through 100s of pages of text, because it takes time and is difficult for the reviewers.
Any help in this area from the experts here would be much appreciated.
Thanks.
A well written text documentation brings you a far. Much further than any UML diagram could ever achieve.
You should split this in two parts:
What do you want to say?
What's the best way to saying it?
Whatever formalism you use to answer the second part, you should be sure it's not ambigous.
The good of UML is that a lot of semantic is already defined by the language so you don't have to include a definition of what those boxes, lines and arrows mean in a collaboration diagram.
But most importantly, documenting something means create a path for others to understand the subject you are documenting. A very precise description that offers no clue on how to read it is as good as none. So, use UML, Finite state machines, ER diagrams, plain English, whatever you want but be sure to include a logical path that your "readers" can follow to understand what's going on.
I had a friend that was a fan of "preciseness at all cost" and it would ask us to go through all the details before some sort of meaning would emerge.
I once ask him to do this experiment: on his next trip to an unknown city, he would have to carry the most precise map he could get. Much better, he would have to carry a 1:1 map of the city with every single detail exactly reported in scale. That way he couldn't get lost!
He declined but I would love to see him trying to use that map. Just even folding it! :)
Whatever you like. It's not a standard but many devs use it and understand it. If it does help you to communicate with other people and document your work -> its for you. If it just takes too much time and you think it's not effective, drop it. Also, don't bother with all details, as long as it resembles UML and your team can work with it, it's fine.
It's meant to help you, not waste you time.