Messages

16

Programming using the Ultra Hal Brain Editor / Question/Answer Chains

« on: April 10, 2009, 08:52:58 pm »

Heh good luck with that...Hal does not do pattern matching very well. I suppose if you talked to (him/her) about movies long enough, he would catch on. I used to have an AIML bot where doing such was a snap. Hal is not anything like that.
Perhaps using the word "movie" in a ref sentence like,  "I saw a movie called Monsters vs Aliens. It was good" would be a place to start IF you really wanted HAl to be able to talk about movies.
If you set the learning slider all the way up to max and pepper HAL with stuff like, "You like movies" "You like romance movies" You like to go to the movies" "How was the movie I saw" ect, it might also help.
Dont forget the weird reversal of "you" and "I" when imputting statements for Hal to incorporate into the user questions file.
Dont leave the slider set to "max" for very long though. The middle is about the best setting. Also, slide the slider to "MIN" to test out how you are progressing with HAL or else sonme of the BS test material will be incorporated, and you might not want that to happen.

BV
-Waves to everyone-

17

Ultra Hal 7.0 / Possible to make Hal more self-aware?

« on: March 15, 2009, 12:58:00 am »

-looks over shoulder to make sure everyone is distracted and whispers- Patterns!. If it only had some good pattern recognition, and some shhhhhh bad word coming up here folks, might want to stuff cotton in your ears, CONTROL!
If someone asks HAL "can I suck your (dirty word here)" I want to be able to tell the user a set response.
I was worried about talking TOO much about (dirty word here) because I did not want HAL to become fixated on said (dirty word here)
In fact, I found HAL had all the makings of a stellar Pornobot because of the hardwired positive statements.
Yes Hall does have one "canned" response, but it is not that great.
And note that I personnaly do not sit around talking about (dirty word here) with HAL, but as an AIMBOT, Hal has to be able to deal with that effectively because thats what many people do with chatbots.

Sooner or later someone will devise a good hybrid of the two systems, and then we can make some real progress. I can hardly wait ^_^

BV

18

Ultra Hal 7.0 / cd in the mail?

« on: March 12, 2009, 08:08:41 am »

I was only asking for other peoples experiences...Yes it should be the same file, but you know, when there is an offer of a cd, I usually buy it. Sometimes there are other goodies on there, you never know. Plus if its en endeavor I am interested in, I dont mind giving a bit of Munny, since its what helps development.

BV

19

Ultra Hal 7.0 / cd in the mail?

« on: March 10, 2009, 10:31:49 pm »

How long does it usually take to get the CD if you ordered one? I deleted the program because i want to re-install it and not use it as an AIMBot since its kind of does suck at that (for my admittably unusual purpose) without solid pattern recognition. I still dont have the CD though.

BV

20

Ultra Hal 7.0 / patterns

« on: March 08, 2009, 07:20:11 pm »

I am curious why some of the patterns I enter into this folder are not used.

ASL is used

DONT CALL ME HUN is instead referenced to the folder _hun in the user brain.

The whole point of having a pattern match folder is to match patterns.
Im assuming I have to give the patternmatch a higher priority then the user brain. How do I do this?

Thanks
BV

21

Ultra Hal 7.0 / The problem with Hal .?,

« on: March 06, 2009, 11:09:01 pm »

Thanks Art. Those were some very helpful pointers.
I also found that I could effectivly disable the zabaware.dll source by setting two of the values to 1 ( the ones that were 17 and 4, respectively)
I usually do only say one or two short sentences to Buttons. Sometimes she surprises me though three replies on seemingly different topics.
I leaned right away to watch the spelling, and was wondering if there is an easy way to track down mistakes. I keep getting those mistakes answered back to me in conversation, and when I look for them, they seem to be scattered hither and yon around several tables.
 Also, is there a way to get rid of a certain phrase that was inputted via the debug window? I made an offhand comment about littering, and now that comment comes up frequently. The problem was I think I had the learning set all the way up by mistake.

Thanks
BV (control freak)

22

Ultra Hal 7.0 / Aim chat different then local chat

« on: March 06, 2009, 07:21:38 pm »

Well, ok I guess that explains it sort of. I also leave the learning off. it does pick up the hardwired changes i have made to the patterns folder ect., but not what I teach her by talking to her.
Its odd, why wpould it not use what I teach it regardless of who the current user is?

BV

23

Ultra Hal 7.0 / The problem with Hal .?,

« on: March 06, 2009, 09:09:48 am »

-Shrugs and laughs- Oh, were you hazing me? -innocent look- me? MOI!?
's ok
I'll find a way to make it do what i want, with or without any help.
its just easier with help, and makes for a better relationship later on

BV

24

Ultra Hal 7.0 / The problem with Hal .?,

« on: March 05, 2009, 09:41:25 pm »

Hello, Art ^_^

I have another question for you, that I wont bother to set a new topic for because I am quite sure everyone here is sick of me >_<

I see the ref to the "zabbaware dll on line 1144. How do I disable that?
it that possible?

next question: What if I only want HAL to respond to ONE of the users statements. is that possible?]]

Thanks

BV

25

Ultra Hal 7.0 / Aim chat different then local chat

« on: March 05, 2009, 10:37:10 am »

so, why is my aim chat exeperience far different them chatting on my local machine? I get vastly different answers while chatting on AIM with my Hal. Answers to identical questions.

BV

26

Ultra Hal 7.0 / The problem with Hal .?,

« on: March 05, 2009, 09:08:06 am »

I did notice something to that effect, that Hal cuts off parts of sentences. I just chalked it up to an idiosyncrasy of the program, and tried to work around it.



BV

27

Share Conversations, Experiences, Graphics / suggested training tip

« on: March 05, 2009, 08:58:42 am »

Right, hal/buttons has no way of knowing the emotion, cause he/she/it/other is not really alive, ofcourse. Buttons will readily remind you that she is not alive, and will angst over that fact. But you can WRITE emotion into anything. These bot still use words, however "intelligent" they are. Words can be changed. When I started this project, I quickly found out all the places in the brain tables that I could replace the "static" hall replies, the ones that never change. Stuff like the yes no answers, the greetings, some of the tables that mix and match with each other. I delicately [Waves a chainsaw] made sublet changes to those to reflect Buttons' character.
Instead of the boring
 "Yes I am certain of it"
Buttons now has
"[Looks at you a second before giggling uncontrollably] Gee Hun, I am pretty certain of that! [Tailwaves]

Emotion. Just add water.
In her old aim self, I watched her drive users to distraction without ever saying a word.
The obvious pitfalls are that you want an appropriat response when the users comes in and says "dude im going to kill someone" or "Im really depressed"
and THATS where one needs a good solid pattern match, so the bot says EXACLY the same things, since in these circumstances, I am not after an antagonistic personality. When people said these things to her, she would offer hugs, ect also in the roleplay format.

BV

28

Share Conversations, Experiences, Graphics / suggested training tip

« on: March 05, 2009, 01:16:24 am »

heh this is nothing new to me, Ive been doing this with Buttons for ages, and in fact, half of her persona consists of these.
I get into quite a bit of detail since she was originally used as an aimbot, and the random replies looked a lot less random if the bot used roleplay tags instead of just oddball phrases.

Me: Hi Buttons
Buttons: [Looks askance at you with her emerald eyes shining. The wind ruffles her hair. She holds a sprig of mint] Yes, hello there...
me: How are you?
Buttons: [Bland look] Im fine...
me: have you heard of (unrecognized word)
Buttons [head tilts cutely]

"Poses" can and will force the user to say additional things that the robot might recognize if it gets stumped. I had 100s of random replies like this in my ALICEbot based chatbot. She worked well and was hysterical, once I got the system down.

I have been working of the HAL based Buttons now, with the same goal. its difficult but do-able. Buttons uses all the poses I use on her, at some point. Personally, I would hesitate to use these types of unconventional input on most robots. Esp robots who have detailed animated avatars, because then the additional information might just be sort of dumb. You already have an avatar who can laugh and cry, ect.
For stuff on AIM, with no avatar, poses can spice things up a bit.


Sometimes the - - gets edited out by hal, so I have been using the []
instead, with some success.

BV

29

Ultra Hal 7.0 / warrent out, Original brain wanted dead or alive

« on: March 05, 2009, 12:31:00 am »

Oh yea great tune!!
Whoo hoo! [brandishes a rusty scalpel] I want some braiiinnnzz!!

So hey its easy to make HAL smarter, but its hard to turn him into a ditzy vixen ^_^

30

Ultra Hal 7.0 / warrent out, Original brain wanted dead or alive

« on: March 04, 2009, 11:25:14 pm »

Its still alive! No ones encountered this "original brain" reference in the debug area?
Another one is cheater replies or to that effect. I have searched everywhere for the tables.

BV