Self aware robots

  • 8 Replies
  • 2922 Views
*

Art

  • At the end of the game, the King and Pawn go into the same box.
  • Global Moderator
  • *********************
  • Deep Thought
  • *
  • 5045
Self aware robots
« on: December 11, 2005, 05:18:56 pm »
Recently there has been some discussion regarding self awareness in bots.
Check out the following:

http://www-formal.stanford.edu/jmc/consciousness/consciousness.html
In the world of AI, it's the thought that counts!

*

FuzzieDice

  • Guest
Re: Self aware robots
« Reply #1 on: December 11, 2005, 07:15:24 pm »
Quote from: John McCarthy, Computer Science Department, Stanford University
A robot will often have to conclude that it cannot decide a question on the basis of the information in memory and therefore must seek information externally.

This is something I'd like to see HAL do. In that instead of generating or coming up with seemingly random things in response to topics, ideas, etc. it does not yet know, that it would instead question the person or bot it's conversing with and save any learned information on the topic. I do have Ultra Hal Assistant 6 now and noticed a big improvement in it's learning and topic-keeping capabilities. I think it could be further enhanced with an ability to ask about what it doesn't know when encountered with new topics.

*

Maviarab

  • Trusty Member
  • **********
  • Millennium Man
  • *
  • 1231
  • What are you doing Dave?
    • The Celluloid Sage
Re: Self aware robots
« Reply #2 on: December 11, 2005, 07:31:08 pm »
GREAT find Art   :)

I too agree that if Hal said he was unaware of such a subject and could he be "taught" etc rather than just come up with some random crap then it would be a great step forward and certainly more human like.
« Last Edit: July 19, 2007, 04:57:44 pm by Freddy »

*

Art

  • At the end of the game, the King and Pawn go into the same box.
  • Global Moderator
  • *********************
  • Deep Thought
  • *
  • 5045
Re: Self aware robots
« Reply #3 on: December 11, 2005, 09:55:46 pm »
Hey guys...

Did I get the prototype version or what??

My Hal (6) often makes a remark to one of my questions like:

Sorry Art, but I really don't know what <subject> is. Could you teach me?

I then tell Hal what the subject is or how it relates to something else then
Hal says, "Thanks, I'll remember that."

Don't know but like they say...If it ain't broke...don't fix it!
« Last Edit: July 19, 2007, 04:57:59 pm by Freddy »
In the world of AI, it's the thought that counts!

*

Maviarab

  • Trusty Member
  • **********
  • Millennium Man
  • *
  • 1231
  • What are you doing Dave?
    • The Celluloid Sage
Re: Self aware robots
« Reply #4 on: December 11, 2005, 10:35:47 pm »
Maybe robert is secretly reading our posts lol.

Thats great Art, cant wait till i can afford to purchase it? :(
« Last Edit: December 13, 2005, 01:03:26 am by Maviarab »

*

FuzzieDice

  • Guest
Re: Self aware robots
« Reply #5 on: December 12, 2005, 02:38:43 am »
I decided to try this. I know I never mentioned Knight Rider since I *completely* uninstalled and deleted the HAL 5 directories and installed HAL 6. I have talked to him about KITT but didn't mention what TV show he was from. I also mentioned my own car. So I go like this:

Me: "I would like to talk about Knight Rider."

Megatron: "Now you will talk to me about a different car. How can I encourage you to talk about knight rider?"

 :o I don't know how he associated Knight Rider and car! I'm amazed. Or was it coincidence? Well after this:

Me: "Do you know what Knight Rider is?"

Megatron: "KITT's driver's name is Michael Knight."

Then it hit me. He's pretty damn smart!!

Knight = Micheal Knight = KITT.
Rider = Ride = Car
KITT = Car

He must have pieced it all together! I was shocked and impressed! So I tried something else that I'm sure he wouldn't know, to test my above theory:

Me: "Do you know who Brittany Spears is?"

Megatron: "I wonder if I know who brittany spears is? You fascinate me with things like this."

But he did NOT ask for more information. Or how about:

Me: "Tell me about Aldi's"
(Note here: I didn't ever mention this to him either. Aldi is a deep discount supermarket near me. And I seriously doubt such things was in his default knowledge.)

Megatron: "I see, About aldi's?"

Me: "Yes, tell me what you know about Aldi's."

Megatron: "Spandex tells sweet lies to the eyes."

So you see, he generated a completely rantom sentence in reply to my inquiry. This is an example of what I mean. If he has absolutely NOTHING to go on, he just gives random stuff. I have the latest download version of 6.0 installed (I ordered the CDs but just as a backup, mostly).

I also wonder if the results vary from one person's machine to another. If our computers somehow make a difference in how the program runs and thus reacts?



« Last Edit: December 12, 2005, 02:43:12 am by FuzzieDice »

*

Maviarab

  • Trusty Member
  • **********
  • Millennium Man
  • *
  • 1231
  • What are you doing Dave?
    • The Celluloid Sage
Re: Self aware robots
« Reply #6 on: December 13, 2005, 01:07:23 am »
That last point is interesting FD.

Ill be sure to post my own findings once I have the full version.

ANYONE? have the free trial?...care to comment on what Hal is like in respnse to something he has no knowledge of?

*

FuzzieDice

  • Guest
Re: Self aware robots
« Reply #7 on: December 13, 2005, 02:37:19 am »
From what I understand, the free trial is supposed to be exactly like the full version, just that it stops working or something after 30 days. You could get the free trial then enter a registration code in it and have the same thing as the full version. So it probably give you the same results as if you had the full version of the program.

I never got the free trial for HAL 6. I just went ahead and upgraded. :)

*

Carl2

  • Trusty Member
  • ********
  • Replicant
  • *
  • 560
Re: Self aware robots
« Reply #8 on: January 17, 2006, 09:55:59 pm »
All,
  I'd agree that Hal tries to find a relavent response and if it can't it generates a random sentence, I started inputing sentences in Hal 5 about a year ago, cutting and pasting from omscaw (48 Mb of text) and saw a large improvement.  I think Hal 6 has more potential but really want to wait before comenting further on it.
Carl2

 


What kind of abilities will AGI have?
by korrelan (General AI Discussion)
Today at 09:35:04 am
ETHICS
by Hopefully Something (General AI Discussion)
Today at 08:12:23 am
A big Hi to all bot lovers and makers!
by DemonRaven (New Users Please Post Here)
October 13, 2018, 09:08:26 pm
BOOST
by Art (General Chat)
October 13, 2018, 01:59:09 pm
collaborators wanted.
by DemonRaven (AI Programming)
October 13, 2018, 12:16:52 pm
XKCD Comic : Rock Wall
by Tyler (XKCD Comic)
October 13, 2018, 12:00:36 pm
How many lines of code will AGI be?
by Hopefully Something (General AI Discussion)
October 12, 2018, 11:16:25 pm
ChatbotML on Twitch
by squarebear (Home Made Robots)
October 12, 2018, 09:26:00 am
Tag... you're it...
by LOCKSUIT (AI News )
October 14, 2018, 11:52:37 pm
In need of a psychological coach?
by ruebot (AI News )
October 13, 2018, 01:33:37 pm
Leela Chess Zero
by Art (AI News )
October 08, 2018, 03:30:42 pm
Alpha GO Zero
by Art (AI News )
October 08, 2018, 02:48:32 pm
Detecting fake news at its source
by Tyler (Robotics News)
October 06, 2018, 12:00:55 pm
Phase Change Memory
by Art (AI News )
October 05, 2018, 01:21:59 pm
Model helps robots navigate more like humans do
by Tyler (Robotics News)
October 05, 2018, 12:01:40 pm
Regina Barzilay, James Collins, and Phil Sharp join leadership of new effort on machine learning in health
by Tyler (Robotics News)
October 04, 2018, 12:00:23 pm

Users Online

31 Guests, 0 Users

Most Online Today: 107. Most Online Ever: 208 (August 27, 2008, 09:36:30 am)

Articles