[agi] Connectionists: ANNOUNCE: PASCAL Visual Object Classes Recognition Challenge 2007
- Forwarded message from Chris Williams [EMAIL PROTECTED] - From: Chris Williams [EMAIL PROTECTED] Date: Mon, 30 Apr 2007 18:10:41 +0100 (BST) To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED], John Winn [EMAIL PROTECTED], [EMAIL PROTECTED], Mark Everingham [EMAIL PROTECTED] Subject: Connectionists: ANNOUNCE: PASCAL Visual Object Classes Recognition Challenge 2007 PASCAL Visual Object Classes Recognition Challenge 2007 We are running a third PASCAL Visual Object Classes Recognition Challenge. This time there are more classes (twenty), more challenging images, and the possibility of more confusion between classes with similar visual appearance (cars/bus/train, bicycle/motorbike). As before, participants can recognize any or all of the classes, and there are classification and detection tracks. There are also two taster competitions, on pixel-wise segmentation and on person layout (detecting head, hands, feet). The development kit (Matlab code for evaluation, and baseline algorithms) and training data is now available at: http://www.pascal-network.org/challenges/VOC/voc2007/index.html where further details are given. The timetable of the challenge is: * April 2007: Development kit and training data available. * 11 June 2007: Test data made available. * 17 Sept 2007, 11pm GMT: DEADLINE for submission of results. * 15 October 2007: Visual Recognition Challenge workshop (Caltech 256 and PASCAL VOC2007) to be held as part of ICCV 2007 in Rio de Janeiro, Brazil, see http://www.robots.ox.ac.uk/~vgg/misc/iccv07/ Mark Everingham Luc Van Gool Chris Williams John Winn Andrew Zisserman - End forwarded message - -- Eugen* Leitl a href=http://leitl.org;leitl/a http://leitl.org __ ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org 8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] Trouble implementing my AGI Algorithm
--- a [EMAIL PROTECTED] wrote: Help me with the algorithm. Thank you Dear a for anonymous (are you related to Ben?), Before you worry about whether an AGI should be friendly or selfish or religious, first you have to solve some lower level problems in language, vision, hearing, navigation, etc. You might make some progress in each field but eventually you will run into the problem that you can't fully solve any of the problems without solving all of them. For example, images and sound contain writing and speech, so you need to solve language. Then, in order to communicate effectively with a machine, it must have a world model similar to yours, and a lot of this knowledge comes from the other senses. After you have done that, then the next problem is that you are not building a human. You are building a slave. Its sole purpose is to be useful to humans. A human body is not necessarily the best form for serving this purpose. You might build a robot with 4 arms and wheels for legs and sonar instead of vision. Or it might not have a body at all, or maybe thousands of insect sized robots controlled as one. The problem is that this creature will have a world model that is nothing like yours, and that will make communication difficult. With currently available computers we cope with this problem by inventing new terminology or by using existing words in new ways. For example, we talk about an operating system process as running or sleeping even though it has no legs and does not dream. Then there are other mental states, like running in privileged mode, that have no equivalent in humans. In humans, selfishness and friendliness and religion are secondary goals to our main goal, which like all species, is to propagate our DNA. For example, religion achieves this goal by making taboo any form of sex that does not contribute to making children. Therefore, it is inappropriate to program religion into an AGI whose goal is not reproduction, but to serve humans. In your AGI design, you need to choose an appropriate set of emotions and mental states, inventing new ones as needed. -- Matt Mahoney, [EMAIL PROTECTED] - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] The role of incertainty
This is similar to the thread I was workign on recently about Goals that didnt get quite as far as I would have liked either. 1. For use as testing metrics or for our personal goals of What an AGI should achieve, or what is important these goals or classes of problems should be defined as well as we possibly can. A couple obvious upper level classes are Navigation - in a real or virtual world. Natural Language - speaking, reading, talking Basic Problems Solving - given a simple, problem find a solution. 2. More specific for your AGI, What do you see the virtual pets doing? Specifically as end user functions for the consumer, the selling points you would give them, and how the AGI would help these functions. Is it going to be a rich enough situation in general to display more than just a blocks world type of pet, go fetch me the blue ball over there James Ratcliff Benjamin Goertzel [EMAIL PROTECTED] wrote: On 5/1/07, Mike Tintner [EMAIL PROTECTED] wrote:No, I keep saying - I'm not asking for the odd narrowly-defined task - but rather defining CLASSES of specific problems that your/an AGI will be able to tackle. Well, we have thought a lot about -- virtual agent control in simulation worlds (both pets and humanlike avatars) -- natural language question answering -- recognition of patterns in large bodies of scientific data Part of the definition task should be to explain how if you can solve one kind of problem, then you will be able to solve other distinct kinds. We can certainly explain that re Novamente, but IMO it is not the best way to get across how the system works to others with a technical interest in AGI. It may well be a useful mode of description for marketing purposes, however. ben g - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?; ___ James Ratcliff - http://falazar.com Looking for something... - Never miss an email again! Yahoo! Toolbar alerts you the instant new Mail arrives. Check it out. - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] The University of Phoenix Test [was: Why do you think your AGI design will work?]
Richard, What's the point here? You seem to be just being cussed. You're not really interested in the structure of the sciences, are you? Psychosemiotics, first off, does NOT EXIST - so how cognitive science could already cover it is interesting. It has been mooted vaguely - in a book esp. by Howard Smith: psychosemiotics, defined as the study of how we learn, understand, and use the signs of culture (p. 2), offers a way to understand cognition by examining how humans use signs to make meaning of their everchanging physical and cultural environments (p. 3). I posit a more ambitious formulation, - that it should be esp. about how the structure of sign systems reflects the structure of the human brain. I doubt that you're really into this area, because if you were, you'd have noticed that the structure/ division I use (symbols/ graphics/ images) is not a recognized division. No, this whole area is still virgin territory - if you disagree, point out the research or relevant branch(es) of science. Vis a vis: There is an actual picture tree in the brain -- see above quote from you -- which is a direct, unambiguous description of the position defended by the group associated with Kosslyn. - I take that more seriously, although I am v. confident of my position. Link me to a statement of this position of the group associated with Kosslyn, and I will reply in detail. - Original Message - From: Richard Loosemore [EMAIL PROTECTED] To: agi@v2.listbox.com Sent: Friday, May 04, 2007 3:14 PM Subject: Re: [agi] The University of Phoenix Test [was: Why do you think your AGI design will work?] Mike Tintner wrote: Er Richard, you are opening too many too large areas - we could be here till the end of the month. It seems to me you are using language rather loosely at times - inevitably you are going to have problems with what I am saying. If I say psychoSEMIOTICS, for example, that's exactly what I mean, and it's v. different from psychoLINGUISTICS. The latter is concerned with how LANGUAGE use reflects brain/mind structures - the former would obviously be concerned with how ALL sign systems' use reflect mind structures - including all symbolic systems, (words, numbers, morse other codes, programminglanguages), all graphic systems, (maps, icons, cartoons, geometry etc. etc) and all image systems (photographs, videos, statues, 3D-models, etc). - and why our total body of sign systems keeps evolving along certain lines. The way you have just defined it, psychoSEMIOTICS is no different than cognitive science/AI. If it is different, specify how (that was my original question). Re Kosslyn etc, my basic concern is not so much with the relative merits of different sign systems - of language vs images - but of how the brain actually processes information - of what it does to make sense of words and numbers - how it actually works, when you read this text for instance. There is an actual picture tree in the brain, I would suggest - it processes information on at least three levels simultaneously (and not just as it may appear to, on just one). The immediate point here is that this whole area has NOT been covered before by Kosslyn or anyone else (although there may be odd allusions in some places). You wouldn't have had all the arguments we had about this area, if it had been covered. I addressed the arguments you were actually having at the time, which were all focussed on statements like There is an actual picture tree in the brain -- see above quote from you -- which is a direct, unambiguous description of the position defended by the group associated with Kosslyn. If you are interested in the more general issue of how the brain actually processes information, regardless of whether it uses images to do so or not, then welcome to the club: but THAT question is cognitive science, and it is not the same as the question of whether the brain does so using picture trees. Re embodied cognition, you'll just have to look it up - it's a still growing field, still contentious. Eh? I am a cognitive psychologist/cognitive scientist, Mike. I asked you where this growing field is, because I don't see any sign of it. I would be happy to look it up if you would point to it. - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?; -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.5.467 / Virus Database: 269.6.2/785 - Release Date: 02/05/2007 14:16 - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] The role of incertainty
2. More specific for your AGI, What do you see the virtual pets doing? Specifically as end user functions for the consumer, the selling points you would give them, and how the AGI would help these functions. Is it going to be a rich enough situation in general to display more than just a blocks world type of pet, go fetch me the blue ball over there James Ratcliff Well, it's a commercial project so I can't really talk about what the capabilities of the version 1.0 virtual pets will be. But the idea will be to start simple, and then incrementally roll out smarter and smarter versions. And, the idea is to make the pets flexible and adaptive learning systems, rather than just following fixed behavior patterns. One practical limitation is that we need to host a lot of pets on each server... However, we can do some borg mind stuff to work around this problem -- so that each pet retains its own personality, yet benefits from collective learning done basis on the totality of all the pets' memories... -- Ben - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] The role of incertainty
Ben Goertzel writes: Well, it's a commercial project so I can't really talk about what the capabilities of the version 1.0 virtual pets will be. I did spend a few evenings looking around Second Life. From that experience, I think that virtual protitutes would be a more profitable product :) - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] The University of Phoenix Test [was: Why do you think your AGI design will work?]
Mike, Richard is not being difficult. He is trying to ascertain the basis for your beliefs (and get pointers to it). Only from this e-mail did *I* ascertain that you believe that you had made up psychosemiotics. Previously, it looked to me as if you thought you were pointing at established science -- and in that context, Richard's questions were more than reasonable. Further, your statement that Psychosemiotics, first off, does NOT EXIST - so how cognitive science could already cover it is interesting just shows *your* ignorance. Just because you believe that you've invented something new and attached a name to it doesn't mean that it isn't already established science under a different name. I am quite aware of Richard's background and can assure you that you are extremely unlikely to be correct when you're trying to correct him on something in basic cognitive science (especially since you clearly lack a solid grounding in the field). So let me repeat the ending of my last e-mail -- We want to welcome new members to this group but your assumptions and communications style are not making it easy for us (and hopefully, you can recognize the time and effort spent bringing you up to speed). A total novice debating an expert may be a great experience for the novice but does *very* little for the group as a whole except expend time and attention (since the novice is very unlikely to contribute to the expert's understanding until he gets up to speed). I would suggest that it would be most effective if you would adopt a course of LEARNING what the group believes and how it communicates FIRST and DEBATING LATER (after you both have something to debate about *and* the ability to effectively communicate it). As a first step, why don't you try asking specific questions rather than being insulting? Mark - Original Message - From: Mike Tintner [EMAIL PROTECTED] To: agi@v2.listbox.com Sent: Friday, May 04, 2007 11:19 AM Subject: Re: [agi] The University of Phoenix Test [was: Why do you think your AGI design will work?] Richard, What's the point here? You seem to be just being cussed. You're not really interested in the structure of the sciences, are you? Psychosemiotics, first off, does NOT EXIST - so how cognitive science could already cover it is interesting. It has been mooted vaguely - in a book esp. by Howard Smith: psychosemiotics, defined as the study of how we learn, understand, and use the signs of culture (p. 2), offers a way to understand cognition by examining how humans use signs to make meaning of their everchanging physical and cultural environments (p. 3). I posit a more ambitious formulation, - that it should be esp. about how the structure of sign systems reflects the structure of the human brain. I doubt that you're really into this area, because if you were, you'd have noticed that the structure/ division I use (symbols/ graphics/ images) is not a recognized division. No, this whole area is still virgin territory - if you disagree, point out the research or relevant branch(es) of science. Vis a vis: There is an actual picture tree in the brain -- see above quote from you -- which is a direct, unambiguous description of the position defended by the group associated with Kosslyn. - I take that more seriously, although I am v. confident of my position. Link me to a statement of this position of the group associated with Kosslyn, and I will reply in detail. - Original Message - From: Richard Loosemore [EMAIL PROTECTED] To: agi@v2.listbox.com Sent: Friday, May 04, 2007 3:14 PM Subject: Re: [agi] The University of Phoenix Test [was: Why do you think your AGI design will work?] Mike Tintner wrote: Er Richard, you are opening too many too large areas - we could be here till the end of the month. It seems to me you are using language rather loosely at times - inevitably you are going to have problems with what I am saying. If I say psychoSEMIOTICS, for example, that's exactly what I mean, and it's v. different from psychoLINGUISTICS. The latter is concerned with how LANGUAGE use reflects brain/mind structures - the former would obviously be concerned with how ALL sign systems' use reflect mind structures - including all symbolic systems, (words, numbers, morse other codes, programminglanguages), all graphic systems, (maps, icons, cartoons, geometry etc. etc) and all image systems (photographs, videos, statues, 3D-models, etc). - and why our total body of sign systems keeps evolving along certain lines. The way you have just defined it, psychoSEMIOTICS is no different than cognitive science/AI. If it is different, specify how (that was my original question). Re Kosslyn etc, my basic concern is not so much with the relative merits of different sign systems - of language vs images - but of how the brain
Re: [agi] The role of incertainty
Second Life also has a teen grid, by the way, which is not very active right now, but which virtual pets could enhance significantly. Virtual prostitutes are not in the plans anytime soon ;-) On 5/4/07, Derek Zahn [EMAIL PROTECTED] wrote: Ben Goertzel writes: Well, it's a commercial project so I can't really talk about what the capabilities of the version 1.0 virtual pets will be. I did spend a few evenings looking around Second Life. From that experience, I think that virtual protitutes would be a more profitable product :) - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?; - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] The role of incertainty
Is there any already existing competition in this area - virtual adaptive pets - that we can look at? - Original Message - From: Benjamin Goertzel To: agi@v2.listbox.com Sent: Friday, May 04, 2007 4:46 PM Subject: Re: [agi] The role of incertainty 2. More specific for your AGI, What do you see the virtual pets doing? Specifically as end user functions for the consumer, the selling points you would give them, and how the AGI would help these functions. Is it going to be a rich enough situation in general to display more than just a blocks world type of pet, go fetch me the blue ball over there James Ratcliff Well, it's a commercial project so I can't really talk about what the capabilities of the version 1.0 virtual pets will be. But the idea will be to start simple, and then incrementally roll out smarter and smarter versions. And, the idea is to make the pets flexible and adaptive learning systems, rather than just following fixed behavior patterns. One practical limitation is that we need to host a lot of pets on each server... However, we can do some borg mind stuff to work around this problem -- so that each pet retains its own personality, yet benefits from collective learning done basis on the totality of all the pets' memories... -- Ben -- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?; -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.5.467 / Virus Database: 269.6.2/785 - Release Date: 02/05/2007 14:16 - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] The role of incertainty
On a less joking note, I think your ideas about applying your cognitive engine to NPCs in RPG type games (online or otherwise) could work out really well. The AI behind the game entities that are supposedly people is depressingly stupid, and games are a bazillion-dollar business. I hope your business direction works out well for you! - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] rule-based NL system
Hi James, I'm going to handle your questions in reverse order . . . . Do you think learning is a requirement for understanding, or intelligence? Yes, I believe that learning is a requirement for intelligence. Intelligence is basically how fast you learn. Zero learning equals zero intelligence. a reservation serivce has a world model as well, it knwo about 1000+ airline routes and times, it talks to you, saves your preferences for outgoign flight, and can use that to think and come up with a suggestion for an incoming flight, and which airline to take A reservation service does indeed have a world model but it is a *very* simple model with very few object types, relationships, and actions. The 1000+ airline routes and times are merely data within the model and even if they numbered a million they would not increase the size of the *model*. But the most important thing is that the model is absolutely fixed -- i.e. the system doesn't learn. and an expert system as having more intelligence due to a richer world model and more ability to give answers. I would say that the expert system is more capable but would disagree that it has more intelligence (unless it has some sort of learning functionality). If we took a 10 year old child, and stopped their ability to learn, they would still have the ability to do all the things they did before, can go to the store, and play and fix breakfast etc. Again, I would phrase this as the child still has their old capabilities but their intelligence has dropped to zero -- because realistically, they would not maintain the ability to do all the things they did before. Initially, yes -- BUT -- slowly and surely, as their environment changed, they would be less and less capable of dealing with it as they couldn't learn what they needed to cope with the change. But understanding itself doesnt have any special requirement that it understand New things, just the things that are currently considering. Have you seen the things that you're currently considering before? If so, how is rote memorization different from understanding? Mark - Original Message - From: James Ratcliff To: agi@v2.listbox.com Sent: Friday, May 04, 2007 11:24 AM Subject: Re: [agi] rule-based NL system Two problems unfortunatly arise quickly there, 1. Internal World Model. An intelligence must have some form of internal world model, because this is what it operates on internally, its memory, People have a complex world model including everythign we have built up over years, but a reservation serivce has a world model as well, it knwo about 1000+ airline routes and times, it talks to you, saves your preferences for outgoign flight, and can use that to think and come up with a suggestion for an incoming flight, and which airline to take. If the system contains weather data as well, and can use it, then it could be more intelligent. It has a world model built up there, not as complex, but defintly there, and I would rate that as having some level of intelligence and an expert system as having more intelligence due to a richer world model and more ability to give answers. 2. Learning. Probably a contreversial point here, but Do you think learning is a requirement for understanding, or intelligence? For an intelligence, I dont believe it is. If we took a 10 year old child, and stopped their ability to learn, they would still have the ability to do all the things they did before, can go to the store, and play and fix breakfast etc. Now for an AGI to grow and be able to do more and more things, it needs to have the ability to learn. But understanding itself doesnt have any special requirement that it understand New things, just the things that are currently considering. Jame Ratcliff Mark Waser [EMAIL PROTECTED] wrote: What definition of intelligence would you like to use? Legg's definition is perfectly fine for me. How about the answering machine test for intelligence? A machine passes the test if people prefer talking to it over talking to a human. For example, I prefer to buy airline tickets online rather than talk to a travel agent. To pass the answering machine test, I would make the same preference given only voice communication, even if I know I won't be put on hold, charged a higher price, etc. It does not require passing the Turing test. I may be perfectly aware it is a machine. You may substitute instant messages for voice if you wish. What does being preferred by humans have to do with (almost any definition of) intelligence? If you mean that it can solve any problem (i.e. tell a caller how to reach any goal -- or better yet even, assist them) then, sure, it works for me. If it's only dealing with a limited domain, like being a travel agent, then I'd
Re: [agi] The role of incertainty
However, we can do some borg mind stuff to work around this problem -- so that each pet retains its own personality, yet benefits from collective learning done basis on the totality of all the pets' memories... Nice! - Original Message - From: Benjamin Goertzel To: agi@v2.listbox.com Sent: Friday, May 04, 2007 11:46 AM Subject: Re: [agi] The role of incertainty 2. More specific for your AGI, What do you see the virtual pets doing? Specifically as end user functions for the consumer, the selling points you would give them, and how the AGI would help these functions. Is it going to be a rich enough situation in general to display more than just a blocks world type of pet, go fetch me the blue ball over there James Ratcliff Well, it's a commercial project so I can't really talk about what the capabilities of the version 1.0 virtual pets will be. But the idea will be to start simple, and then incrementally roll out smarter and smarter versions. And, the idea is to make the pets flexible and adaptive learning systems, rather than just following fixed behavior patterns. One practical limitation is that we need to host a lot of pets on each server... However, we can do some borg mind stuff to work around this problem -- so that each pet retains its own personality, yet benefits from collective learning done basis on the totality of all the pets' memories... -- Ben -- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?; - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] The University of Phoenix Test [was: Why do you think your AGI design will work?]
Mike Tintner wrote: Richard, What's the point here? You seem to be just being cussed. You're not really interested in the structure of the sciences, are you? Is this ad hominem remark really necessary? Psychosemiotics, first off, does NOT EXIST - so how cognitive science could already cover it is interesting. It has been mooted vaguely - in a book esp. by Howard Smith: Okay, I'll try to phrase it as carefully as I can: what you suggest as the subject matter of 'psychosemiotics' does not seem to differ from the subject matter of cognitive science/psychology, because the latter already is committed to understanding cognition in all its aspects, including the rather small aspect of cognition that is the human use of signs .. so if you think there is something special about psychosemiotics that makes it distinct from what cognitive science is already doing, please specify this. psychosemiotics, defined as the study of how we learn, understand, and use the signs of culture (p. 2), offers a way to understand cognition by examining how humans use signs to make meaning of their everchanging physical and cultural environments (p. 3). I posit a more ambitious formulation, - that it should be esp. about how the structure of sign systems reflects the structure of the human brain. I doubt that you're really into this area, because if you were, you'd have noticed that the structure/ division I use (symbols/ graphics/ images) is not a recognized division. No, this whole area is still virgin territory - if you disagree, point out the research or relevant branch(es) of science. All you have done so far is to declare that Semiotics should be used to shed light on the structure of the human mind, and that this should be called psychosemiotics, and that this is virgin territory. My response to you is the same as the response I would give to someone who might claim that the human use of restaurants should be used to shed light on the structure of the human mind, and that this should be called psychobistromathics, and that this is virgin territory. I would ask: why is this different from the general use of all kinds of human behaviors to study the mind . a field that is already named, and is called cognitive science? Most people would say that it has to be a good deal more than just a vague declaration of intent, to be a scientific field with a new name. (BTW Someone already did employ the human use of restaurants as a way to shed light on the structure of the human mind, but they were never inclined to declare it a new field of study, or promise, before they had even started on it, that it was a virgin territory). Vis a vis: There is an actual picture tree in the brain -- see above quote from you -- which is a direct, unambiguous description of the position defended by the group associated with Kosslyn. - I take that more seriously, although I am v. confident of my position. Link me to a statement of this position of the group associated with Kosslyn, and I will reply in detail. Try any basic undergraduate text on cognitive science, or, if you are in a hurry, I am sure you will be able to find a statement of their position somewhere in these, or a thousand other places: http://en.wikibooks.org/wiki/Cognitive_Psychology_and_Cognitive_Neuroscience/Imagery http://www.iep.utm.edu/i/imagery.htm http://www.bcp.psych.ualberta.ca/~mike/Pearl_Street/Dictionary/contents/I/imagery.html http://users.ecs.soton.ac.uk/~harnad/Hypermail/Foundations.Cognition/0091.html http://ruccs.rutgers.edu/faculty/pylyshyn-mehler.htm http://mitpress.mit.edu/catalog/item/default.asp?tid=7103ttype=2 http://www.gis.net/~tbirch/mi11.htm http://www-static.cc.gatech.edu/~jimmyd/summaries/kosslyn1994.html - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] What would motivate you to put work into an AGI project?
What would motivate you to put work into an AGI project? 1) A reasonable point of entry into the project 2) The project would need to be FOSS, or at least communally owned. (FOSS for preference.) I've had a few bad experiences where the project leader ended up taking everything, and don't intend to have another. 3) The project would need to be adopting a multiplex approach. I don't believe in single solutions. AI needs to represent things in multiple ways, and to deal with those ways in quasi-independent channels. My general separation is: Goals (desired end states), Desires (desired next states), Models, and logic. I recognize that everything is addressed by a mixture of these approaches...but people seem to use VERY different mixtures (both from person to person and in the same person from situation to situation). 4) I'd need to have a belief that the project had a sparkplug. Otherwise I might as well keep fumbling around on my own. Projects need someone to inspire the troops. 5) There would need to be some way to communicate with the others on the project that didn't involve going to a restaurant. (I'm on a diet, and going to restaurants frequently is a really BAD idea.) (N.B.: One project I briefly joined had a chat list...which might have worked well if it had actually been the means of communication. Turned out that the inner circle met frequently at a restaurant and rarely visited the chat room. But I think a mailing list or a newsgroup is a better choice anyway. [The project was successful, but I think that the members on the chat group were mainly a diversion from the actual work of the project.]) 6) Things would need to be reasonably documented. This comes in lots of forms, but for a work in progress there's a lot to be said for comments inserted into the code itself, and automatically extracted to create documentation. (Otherwise I prefer the form that Python uses...but nobody else does that as well.) 7) LANGUAGES: Using a language that I felt not completely unsuitable. After LOTS of searching I've more or less settled on Java as the only wide-spread language with decent library support that can run distributed systems with reasonable efficiency. There are many other contenders (e.g., C, C++, Fortran, Alice, Erlang, and D each have their points), and I don't really *like* Java, but Java, C, and C++ appear to be the only widely used languages that have the ability to run across a multi-processor with reasonable efficiency. (And even there the techniques used can hardly be called widespread.) 7a) Actually C and C++ can be suitable if there are appropriate libraries to handle such things as garbage collection, and protocols for how to save persistent data and then remember it later. But I still don't like the way they make free use of wild pointers. 7b) I wonder to what extent the entire project needs to be in the same language. This does make understanding things easier, as long as it's small enough that someone can understand everything at a low level, or if the entity should ever want to understand itself. But there are plausible arguments for writing things in a rapid development language, such as Python or Ruby, and then only translating the routines that later need to be translated for efficiency. (If only those languages could execute across multiple processors!) - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] rule-based NL system
J. Storrs Hall, PhD. wrote: On Wednesday 02 May 2007 15:08, Charles D Hixson wrote: Mark Waser wrote: ... Machines will know the meaning of text (i.e. understand it) when they have a coherent world model that they ground their usage of text in. ... But note that in this case world model is not a model of the same world that you have a model of. After reading the foregoing discussions of subjects such as intelligence, language, meaning, etc, it is quite clear to me that the various members of this list do not have models of the same world. This is entirely appropriate: consider each of us as a unit in a giant GA search for useful ways of thinking about reality... Josh Well, that's true. E.g., when I was 3 I had one I patched for 3 months in a vain attempt to cure amblyopia. This caused me to be relatively detached from visual imagery, and more attached to kinesthetic imagery. But still, all normal people have a world model where when their eyes are covered they can't see, but where the eyes cannot be removed and then replaced. So there are relatively small degrees of difference between the world models of normal humans and those which will be learned by AGIs. This is even true in the case of AGIs which are raised with the intention of having them have approximately normal maturation. The attempt is essentially futile. Humans will come to resemble AGIs before AGIs come to resemble people. (Admittedly, though, the AGIs that people eventually come to resemble won't bear much resemblance to the early model AGIs.) - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] The University of Phoenix Test [was: Why do you think your AGI design will work?]
I think at some point in time an AGI will need to be embodied.. I know many intend to use robots in the future, and to copy the software into them, as a step embodiment in a virtual environment could prove useful. One thing I intend to do with mine is give it autonomy as soon as possible, and allow it to explore, and try things out. This is a crucial route in learning, and by letting a bot loose in a sim office environment, it could act and interact with many objects and learn in that fashion. There is only so much we can do with text only, many things will be learned as experienced, with text we can pull in some initial material and much facts, but not all of the life experience. A simplified version of vision can be used, goign ahead and letting a bot know or read the name of many of the objects, to get around the specific vision problems that are known. Second life now has a programming API as well, and I just found a damage function which could be used for role-playing agents. James Ratcliff YKY (Yan King Yin) [EMAIL PROTECTED] wrote: On 4/27/07, James Ratcliff [EMAIL PROTECTED] wrote: *navigation, *manipulation, *communication/ languages You're talking about stepping stone tasks to bootstrap an AGI, so I'd say navigation and manipulation are not necessary. In fact, the whole embodiment thing can be ditched, because language is sufficient. In fact, a robot or VR-bot cannot navigate / manipulate without vision, so the embodiment route is too complex and inefficient compared to the NL route. I'd add: * basic reasoning abilities * memorizing facts and maintain a coherent body of knowledge otherwise the thing would be useless. Agree? YKY - Ahhh...imagining that irresistible new car smell? Check outnew cars at Yahoo! Autos. - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] rule-based NL system
Yeah I am trying to get his to run, but no luck yet, wish it wasnt only linux based, But even a general 3-D graphical app is not that hard to write, I have done a few, and I am also looking at something like using a Second Life interface, as much of the graphics and interface design has already been done, and there is a rich environment and interface there that could be built upon. I also wrote a bot for World of Warcraft, though I dont believe the environment is rich enough for full interactions needed by an AGI. Once you could get to a level of telling the AGI to do something like Fill up that bucket full of water, having it respond with ? Dont know how, please show me, and then being able to use your character to specificially show it how to do tasks, you would be in a good position to have a teachable robot that could then generalize on these tasks to learn how to do many different things. James Ratcliff YKY (Yan King Yin) [EMAIL PROTECTED] wrote: On 5/4/07, James Ratcliff [EMAIL PROTECTED] wrote: The point of most of this is humans and an AI would need to construct a imaginary world environment in their mind. Most people make a typical elephant, and a typical chair and then interact the to as directed. A blind person still gets its information from experience... if it reads about an elephant, it proabbly says a big animal the size of a car, and her experience lets her know abnout cars and animals, and she has sat in chairs and know how big they are. But both of those are tied to the physical experences that she has. You can only get so much from the words alone unless you have an infinite database where everything poeeible has been described fully. But many many things can be gathered from the text alone as well. A VR interface would certainly be nice, but it takes a lot of time to build one and I'm not good at that area. Maybe Ben's AGI-Sim can be used by another AGI? If so we can save a lot of efforts. YKY - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?; ___ James Ratcliff - http://falazar.com Looking for something... - Ahhh...imagining that irresistible new car smell? Check outnew cars at Yahoo! Autos. - This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415user_secret=fabd7936
Re: [agi] rule-based NL system
I would say rote memorization and knowledge / data, IS understanding. OK, we have a definitional difference then. My justification for my view is that I believe that you only *really* understand something when you have predictive power on cases that you haven't directly seen yet (sort of like saying that, in order to be useful or have any value, a hypothesis must have predictive power). I look outside and I see a tree, I understand that it is a tree, I know its a tree, I know about leaves and grass and how it grows... I havnt learned anything new, I memorized all that from books and teaching etc. I don't think so. I think that you have a lot of information that you derived from generalizations, analogies, etc (i.e. learning). I would further say that I given the level of knowledge and understanding about the tree that I was intelligent in that area, you could ask me questions and I could answer them, I could conjecture what would happen if I dug the tree up etc. Are you *sure* that you've been directly told what would happen if you dug a tree up? What do you think would happen if you dug up a planticus imaginus? I'm sure that you haven't been specifically told what would happen then. :-) I think that you have some serious predictive power that is *not* just rote memorization. Learning does not seem to be a requirment for intelligence, though a good intelligence, and a growing intelligence would need to learn. Your definition of intelligence is apparently (and correct me if I'm wrong) how well something deals with it's environment. My contention is that anything that doesn't learn will necessarily undergo a degradation of their ability to deal with it's environment. If you agree with this, then why don't you agree with learning being a requirement for intelligence? Mark - Original Message - From: James Ratcliff To: agi@v2.listbox.com Sent: Friday, May 04, 2007 4:56 PM Subject: Re: [agi] rule-based NL system I would say rote memorization and knowledge / data, IS understanding. I look outside and I see a tree, I understand that it is a tree, I know its a tree, I know about leaves and grass and how it grows... I havnt learned anything new, I memorized all that from books and teaching etc. I would further say that I given the level of knowledge and understanding about the tree that I was intelligent in that area, you could ask me questions and I could answer them, I could conjecture what would happen if I dug the tree up etc. Learning does not seem to be a requirment for intelligence, though a good intelligence, and a growing intelligence would need to learn. James Ratcliff Mark Waser [EMAIL PROTECTED] wrote: Hi James, I'm going to handle your questions in reverse order . . . . Do you think learning is a requirement for understanding, or intelligence? Yes, I believe that learning is a requirement for intelligence. Intelligence is basically how fast you learn. Zero learning equals zero intelligence. a reservation serivce has a world model as well, it knwo about 1000+ airline routes and times, it talks to you, saves your preferences for outgoign flight, and can use that to think and come up with a suggestion for an incoming flight, and which airline to take A reservation service does indeed have a world model but it is a *very* simple model with very few object types, relationships, and actions. The 1000+ airline routes and times are merely data within the model and even if they numbered a million they would not increase the size of the *model*. But the most important thing is that the model is absolutely fixed -- i.e. the system doesn't learn. and an expert system as having more intelligence due to a richer world model and more ability to give answers. I would say that the expert system is more capable but would disagree that it has more intelligence (unless it has some sort of learning functionality). If we took a 10 year old child, and stopped their ability to learn, they would still have the ability to do all the things they did before, can go to the store, and play and fix breakfast etc. Again, I would phrase this as the child still has their old capabilities but their intelligence has dropped to zero -- because realistically, they would not maintain the ability to do all the things they did before. Initially, yes -- BUT -- slowly and surely, as their environment changed, they would be less and less capable of dealing with it as they couldn't learn what they needed to cope with the change. But understanding itself doesnt have any special requirement that it understand New things, just the things that are currently considering. Have you seen the things that you're currently considering before? If so, how is rote memorization different from understanding? Mark - Original Message
Re: [agi] Trouble implementing my AGI Algorithm
I do not believe that the algorithm must be more complex. The more complex the algorithm, the more ad hoc it is. Complex algorithms are not able to perform generalized tasks. I believe the reason that n-digit was a failure was because there is no vision system, NOT because the algorithm is too simple. Because the algorithm searches the database recursively, I believe that my simple algorithm can perform any computation (trained by operant conditioning). The failure for n-digit addition was because there are no eyes that can move to concentrate on each digit. The database is remarkably similar to the human brain. It can learn easily by only remembering the difference between the external stimuli with a similar stimuli remembered in the database. Therefore, the algorithm compress the learned knowledge efficiently. Pattern recognition and abstract reasoning is also easy because of the incremental learning. I am having trouble with the fuzzy database representation. So it's best to test the algorithm in a specific subfield (like n-digit addition) and then generalize it into real-world tasks. In general, my algorithm behaves like the brain of an animal. Animals learn by operant conditioning and are also difficult to teach them multiple digit addition. I believe that the environment must be fuzzy in order for the operant conditioning method to work. I know that the database has to remember pain and pleasure for stimuli. But I have difficulty making a fuzzy database representation, even for some subfields. - Original Message From: Mark Waser [EMAIL PROTECTED] To: agi@v2.listbox.com Sent: Thursday, May 3, 2007 5:06:33 PM Subject: Re: [agi] Trouble implementing my AGI Algorithm Interesting e-mail. I agree with most of your philosophy but believe that the algorithm you are requesting is far, far more complex than you realize. Is there any particular reason why you're remaining anonymous? - Original Message - From: a [EMAIL PROTECTED] To: agi@v2.listbox.com Sent: Thursday, May 03, 2007 4:57 PM Subject: [agi] Trouble implementing my AGI Algorithm Hello, I have trouble implementing my AGI algorithm: The below paragraphs might sound ridiculous, because they are my original ideas. We are all motivated by selfish thoughts. We help others so others can help us back. We help others to cope with our pleasurable chemical addiction. We help others because helpfulness is encoded in our genetic markup. We experience pain. Pain is to help us defend damage. When we touch something hot we can draw back. But we have the free will to not react to it. I believe there is no free will. I will explain what I means. Assume that pain is a constraint. But this constraint is not absolute. Other thoughts can override the constraint. For example, when you help some animal being eaten from a monster, you can fight with the monster to save the animal's life. But you will experience pain in the fight. Therefore pain is not a constraint. Your goal to save the animal's life overrides the pain constraint. (your goal to save the animal's life is also motivated by selfish actions) Therefore, pain is not a constraint. But if there is no goal that overrides the pain constraint, you will do anything to avoid the pain. We have proven there is no free will--we choose to react or not react to pain is dependent on your goal or our knowledge. Therefore, implementing pain as a constraint in friendly AI will not help many lives. Our brains are doing things to get the highest pleasure as possible. We get a chemical addiction to save that animal. That pleasure is more pleasant than avoiding the pain by not fighting. We trust ourselves. We can gamble pain for future pleasure. Therefore, I believe that emotion can be implemented by an ordinary computer. Emotion can be implemented by an algorithm that searches for the highest pleasure. The algorithm must also has the ability to gamble pain for pleasure (by applying goals or knowledge). There is no right or wrong. We kill insects all the time. But we usually do not sympathize with them. This is because that our religion says that bugs are not as important as other animals. It's a byproduct of natural selection. We have to hunt animals to survive. Without religion, we would brood over this question: Is it better to save a human by sacrificing 1000 insects or vice versa? Therefore we assume that religion is natural. Religion helps us survive. Some religions help us believe there is afterlife and reincarnation. Because we believe these, we do not fear death. We are not afraid to sacrifice ourselves for others. For example, we will not be afraid to participate in wars and spread our religion. Religion is a virus. Most of the world is religious because of that. Therefore, some religions are dangerous. But religion is essential for our daily survival. Some religious thoughts are
Re: [agi] rule-based NL system
Its mainly that I believe there is a full range of intelligences available, from a simple thermostat, to a complex one that measures and controls humudity and knows if a person is in a run, and has specific settings for differnt people, to a an expert system, to a human to an AI and super AGI, all having some level of intelligence. The ones we are concerned with are the 1/2 human level and anything above. Learning I would say is a key role in having a high-level of intelligence, probably the main building block, learning and reasoning, both tied tightly together. James Ratcliff Mark Waser [EMAIL PROTECTED] wrote: I would say rote memorization and knowledge / data, IS understanding. OK, we have a definitional difference then. My justification for my view is that I believe that you only *really* understand something when you have predictive power on cases that you haven't directly seen yet (sort of like saying that, in order to be useful or have any value, a hypothesis must have predictive power). I look outside and I see a tree, I understand that it is a tree, I know its a tree, I know about leaves and grass and how it grows... I havnt learned anything new, I memorized all that from books and teaching etc. I don't think so. I think that you have a lot of information that you derived from generalizations, analogies, etc (i.e. learning). I would further say that I given the level of knowledge and understanding about the tree that I was intelligent in that area, you could ask me questions and I could answer them, I could conjecture what would happen if I dug the tree up etc. Are you *sure* that you've been directly told what would happen if you dug a tree up? What do you think would happen if you dug up a planticus imaginus? I'm sure that you haven't been specifically told what would happen then. :-) I think that you have some serious predictive power that is *not* just rote memorization. Learning does not seem to be a requirment for intelligence, though a good intelligence, and a growing intelligence would need to learn. Your definition of intelligence is apparently (and correct me if I'm wrong) how well something deals with it's environment. My contention is that anything that doesn't learn will necessarily undergo a degradation of their ability to deal with it's environment. If you agree with this, then why don't you agree with learning being a requirement for intelligence? Mark - Original Message - From:James Ratcliff To: agi@v2.listbox.com Sent: Friday, May 04, 2007 4:56 PM Subject: Re: [agi] rule-based NLsystem I would say rote memorization and knowledge / data, ISunderstanding. I look outside and I see a tree, I understand that it isa tree, I know its a tree, I know about leaves and grass and how itgrows... I havnt learned anything new, I memorized all that from booksand teaching etc. I would further say that I given the level ofknowledge and understanding about the tree that I was intelligent in thatarea, you could ask me questions and I could answer them, I could conjecturewhat would happen if I dug the tree up etc. Learning does not seem tobe a requirment for intelligence, though a good intelligence, and a growingintelligence would need to learn. James Ratcliff MarkWaser [EMAIL PROTECTED] wrote: Hi James, I'm going to handle your questions in reverse order . . . . Do you think learning is a requirement for understanding, or intelligence? Yes, I believe that learning is a requirement for intelligence. Intelligence is basically how fast you learn. Zero learning equals zero intelligence. a reservation serivce has a world model as well, it knwo about 1000+ airline routes and times, it talks to you, saves your preferences for outgoign flight, and can use that to think and come up with a suggestion for an incoming flight, and which airline to take A reservation service does indeed have a world model but it is a *very* simple model with very few object types, relationships, and actions. The 1000+ airline routes and times are merely data within the model and even if they numbered a million they would not increase the size of the *model*. But the most important thing is that the model is absolutely fixed -- i.e. the system doesn't learn. and an expert system as having more intelligence due to a richer world model and more ability to give answers. I would say that the expert system is more capable but would disagree that it has more intelligence (unless it has some sort of learning functionality). If we took a 10 year old child, and stopped their ability