Virtual Humans Forum
Virtual Humans Forum
Home | Profile | Register | Active Topics | Members | Search | FAQ
Username:
Password:
Save Password
Forgot your Password?

Virtual Humans
Partner website:
Chatbots.org
 All Forums
 Virtual Humans
 Artificial Intelligence
 human mind map
 New Topic  Reply to Topic
 Printer Friendly
Previous Page | Next Page
Author  Topic Next Topic
Page: of 25

Nomeneiste
Intrigued Member



USA
121 Posts

Posted - Sep 28 2009 :  04:08:04  Show Profile  Click to see Nomeneiste's MSN Messenger address  Reply with Quote
Good morning, Sir.

Good luck with the surgery!

- The answer is oorple.
Go to Top of Page

toborman
Hooked Member



USA
291 Posts

Posted - Oct 06 2009 :  00:36:43  Show Profile  Visit toborman's Homepage  Reply with Quote
Survived the surgery.

Completely revised the mind map site to coincide with the diagram.

http://mindmap.iwarp.com
Go to Top of Page

toborman
Hooked Member



USA
291 Posts

Posted - Oct 06 2009 :  22:41:49  Show Profile  Visit toborman's Homepage  Reply with Quote
The Harry site is becoming a place where people can see how an intelligent agent is put together. Eventually you will be able to add you own plugins. Let me know what you would like to see.

http://mindmap.iwarp.com
Go to Top of Page

toborman
Hooked Member



USA
291 Posts

Posted - Oct 15 2009 :  08:49:58  Show Profile  Visit toborman's Homepage  Reply with Quote
Trying to determine a suitable test for self awareness. What do you think?

Tests for self awareness.
· Able to remember, recall and describe sensations felt.
· Able to hear and describe one’s own thoughts.
· Able to remember, recall and describe episodes experienced.

http://mindmap.iwarp.com
Go to Top of Page

GrantNZ
Dedicated Member



New Zealand
2677 Posts

Posted - Oct 15 2009 :  17:49:22  Show Profile  Reply with Quote
Congrats on the surgery I hope it was a success?

Regarding the self-awareness tests, it's a good start but they seem to focus a lot on memory and less on thought.

I haven't put a lot of thought into this, so my ideas are not refined, but I'd include tests like:
  • Able to define one's self in the context of the (known and perceived) external world.
  • Able to recognise one's self (without hints or prompting) from a description, and from observing one's own actions (e.g. in a mirror system, without having had the explicit definition "this is a mirror and shows yourself").
  • Able to understand the effects of the external world on one's self, and (ideally) able to interact with the world to promote a deliberate change in one's own self condition.
  • Able to identify one's own thoughts and understand their action. Able to guess the effects of a potential thought (although not necessarily with accuracy, as that is where self-learning comes in), and use that guess in future analysis and planning. Ideally, able to deliberately modify one's own thought processes and systems to alter one's self condition.
As I say, not refined, and I'd be interested to hear people's improvements/additions/deletions
Go to Top of Page

toborman
Hooked Member



USA
291 Posts

Posted - Oct 17 2009 :  04:50:44  Show Profile  Visit toborman's Homepage  Reply with Quote
good ideas. taking them one at a time.

quote:
Able to define one's self in the context of the (known and perceived) external world.


what would the test look like? how would test maker predict a "correct" response?

http://mindmap.iwarp.com
Go to Top of Page

GrantNZ
Dedicated Member



New Zealand
2677 Posts

Posted - Oct 17 2009 :  08:13:18  Show Profile  Reply with Quote
Good question

Perhaps part one could be the simple question "define yourself," where we might expect an answer such as "I am a computer-based virtual intelligence," however such a phrase should not be simply stored in the AI somewhere - the AI should be using its world database to find that answer. Ideally, as the AI learns about its world and identifies that things can be described with attributes or divided into categories, it should be curious as to which categories and attributes apply to itself, and use that information in its self-definition.

Part two could be simple factual questions and answers - "are you computer-based;" "are you a mammal;" "where do you get energy from;" etc. "Are you alive?"

Part three could look at inferences, to see if the AI has identified things that are part of its being, without this having been explicitly defined in terms of self-awareness. The AI should be able to explain reasons. Things like "do you run on electricity" (which it can infer by knowing that it is computer-based and through information about computers); "do you look like a human;" "what will happen to you if there is a power outage;" "are you able to apply for a passport;" "who created you and why, and are you meeting those goals;" "how do you feel about your existance;" etc.

Part four could focus on comparisons. Human: "How similar are you to a cat?" AI: "We both have feelings, goals, and require sustenance. However the cat is a biological being, and consumes food and oxygen. In comparison, I am computer-based and consume electricity. I have a connection to the internet which the cat does not have." And so on. Other questions: "How similar are you to a human?" "... to UltraHal?" "... to a table?" "... to Fred?" (With the Fred example, looking for recognition of attributes that can be applied within a category - i.e. Fred is a human, but may have characteristics like "irritable" which are not applied to humans in general but can be applied to a specific instance of one.)

This all relies, of course, on the AI having learned all this information previously, but without having been coached specifically on how to answer self-awareness questions. After all, even AIML could answer these things "correctly," if the patterns had been scripted that way. On the other hand, a lot of humans have false beliefs about themselves which have been coached (whether from internal or external influences), such as "I'm ugly," "I'm stupid," etc, which don't hold up under strictly rational and logical analysis but still form a major part of their self-image; so maybe some coaching influence should be allowed

Anyway, again, my thoughts aren't refined or anything, I'm just throwing out ideas in the hope of stimulating discussion and consensus (where possible )
Go to Top of Page

toborman
Hooked Member



USA
291 Posts

Posted - Oct 17 2009 :  11:09:34  Show Profile  Visit toborman's Homepage  Reply with Quote
quote:
Perhaps part one could be the simple question "define yourself," where we might expect an answer such as "I am a computer-based virtual intelligence," however such a phrase should not be simply stored in the AI somewhere - the AI should be using its world database to find that answer. Ideally, as the AI learns about its world and identifies that things can be described with attributes or divided into categories, it should be curious as to which categories and attributes apply to itself, and use that information in its self-definition.


Test 1 for self awareness.
Given: kari is a conversationalist.
Given: kari is a computer program.
Given: tom is a conversationalist.
Given: tom is a human.
Given: you are a conversationalist.
Q: describe yourself.
A: I am a conversationalist.
A: I am possibly a human or possibly a computer program.

Is this test adequate?

http://mindmap.iwarp.com
Go to Top of Page

toborman
Hooked Member



USA
291 Posts

Posted - Oct 21 2009 :  06:40:11  Show Profile  Visit toborman's Homepage  Reply with Quote
quote:
I am possibly a human or possibly a computer program.



or possibly a human computer program.
quote:
"I am a computer-based virtual intelligence"

http://mindmap.iwarp.com
Go to Top of Page

toborman
Hooked Member



USA
291 Posts

Posted - Apr 28 2010 :  10:51:28  Show Profile  Visit toborman's Homepage  Reply with Quote
added language acquisition process to mindmap site and here for your comments.

Language acquisition is the process of associating mental representations with oral, written, graphic and gestured communication in an agreed upon mutual language that allows comprehension and expression. This process includes accepting communication and propositions that express intent, inferring grammar and style rules, and testing the rules.

The following is a “learn by example” English language acquisition process in a simple sentence written dialog protocol.

1. Using a known sentence structure, parse the input sentence
2. Compare components to English language patterns
o Case 1 - If 100% pattern match found, then exit (to interpret sentence)
o Case 2 - If literals only match found, then
1. For each unmatched element, infer the class of the element is probably consistent with the pattern.
2. Exit (to interpret sentence)
o Case 3 - If no matches found, then for each stored sentence in past dialogs 1. Parse sentence into words and delimiters using sentence structure
2. Compare input components to dialog sentence components
3. If match not found, then go to next dialog sentence
4. If match found, then infer pattern for related sentences using matched components using inductive generalization
5. Compare non-matching component to class elements
6. If match found, then go to compare next non-matching component
7. If no match found, then establish a class variable for the non-matching component
8. Remember class variables
9. Remember English language pattern
10. Exit (to interpret sentence)
o Case 4 - If variables only found, then
1. Assume unmatched components are new literal
2. Create probable new pattern
3. Remember English language pattern
4. Exit (to interpret sentence)

Case 1 – 100% match.
Input: Mary went to the doctor’s office.
Dialog entries: Mary went to the store. Alice went to the doctor’s office.
English Language Pattern: person went to the place.
Class: person (Mary, Alice)
Class: place (store, doctor’s office)
Sentence structure (words + delimiter)

Case 2 – literals only match.
Input: Jake went to the gym.
Dialog entries: Mary went to the store. Alice went to the doctor’s office.
English Language Pattern: person went to the place.
Class: person (Mary, Alice)
Class: place (store, doctor)
Sentence structure (words + delimiter)
New elements: Assume that Jake is probably a person and gym is probably a place.

Case 3 – no match.
Input: Jake went to the gym.
Dialog entries: Mary went to the store. Alice went to the doctor’s office.
English Language Pattern: none
Class: person (Mary, Alice)
Class: place (store, doctor’s office)
Sentence structure (words + delimiter)
New elements: Assume that Jake is probably a person and gym is probably a place.
New elements: Assume that “person went to the place” is a new pattern.

Case 4 – variables match.
Input: Mary went to the doctor’s office.
Dialog entries: Mary went to the store. Alice went to the doctor’s office.
English Language Pattern: person was at the place.
Class: person (Mary, Alice)
Class: place (store, doctor’s office)
Sentence structure (words + delimiter)
New elements: Assume that “person went to the place” is a new pattern.

http://mindmap.iwarp.com

Edited by - toborman on Apr 28 2010 10:56:34
Go to Top of Page

kenkirkland
Curious Member



58 Posts

Posted - Apr 28 2010 :  17:02:58  Show Profile  Reply with Quote
Good to see you posting again Toborman!. Things have been two-sided around here ;) I hope to get a snippet of Harry to think about.

Ken
Go to Top of Page

toborman
Hooked Member



USA
291 Posts

Posted - Apr 28 2010 :  17:34:01  Show Profile  Visit toborman's Homepage  Reply with Quote
quote:
Good to see you posting again Toborman!. Things have been two-sided around here ;) I hope to get a snippet of Harry to think about.


thanks ken, it's good to be back. I've been working on an AI book and discovered I that Harry still hasn't learned about analysis, synthesis, creativity, language acquisition, discovery by exploration, set theory and mathematics. so I'm looking forward to discussing these with my friends here.

http://mindmap.iwarp.com
Go to Top of Page

mikmoth
Moderator



USA
2082 Posts

Posted - Apr 28 2010 :  20:22:58  Show Profile  Visit mikmoth's Homepage  Reply with Quote
Hey Taborman. I was just thinking about you the other day. Glad you're still out there. Keep us updated.

 http://lhandslide.com
Go to Top of Page

TheMikh28
Curious Member



USA
39 Posts

Posted - Apr 28 2010 :  21:01:56  Show Profile  Reply with Quote
Welcome back, toborman.

I've been thinking a lot about language acquisition through pattern matching myself, particularly through cross-comparison of "binary" combinations of word sequences. The algorithm I stumbled across and have come to use quite frequently in this area has proved quite helpful; I briefly detailed how it works in my topic.

[M*P]
Go to Top of Page

toborman
Hooked Member



USA
291 Posts

Posted - Apr 28 2010 :  22:10:26  Show Profile  Visit toborman's Homepage
quote:
I've been thinking a lot about language acquisition through pattern matching myself, particularly through cross-comparison of "binary" combinations of word sequences. The algorithm I stumbled across and have come to use quite frequently in this area has proved quite helpful; I briefly detailed how it works in my topic.


I remember your binary approach and weightstack description from the fadela project. Have these approaches been fruitful for your acquisition program?

http://mindmap.iwarp.com
Go to Top of Page
Page: of 25  Topic Next Topic  
Previous Page | Next Page
 New Topic  Reply to Topic
 Printer Friendly
Jump To:
Virtual Humans Forum © V.R.Consulting Go To Top Of Page
This page was generated in 0.16 seconds. Snitz Forums 2000