RE: [agi] Thought experiment on informationally limited systems
Is an AGI necessarily a super human or is just the equivalent of a smart human good enough? I rarely find humans that adapt by itself to unfamiliar situations. Take a random person and place them in some remote location without adequate training and they would last how long? A week maybe. I wouldn't define this as adapting very well to an unfamiliar situation, would you? Most humans (99.999% IMHO) don't ever create anything absolutely new, in terms of human knowledge, so why would this be a criteria for an AGI? I agree that an AGI must be able to learn. I agree an AGI must be able to reason and solve problems without just resorting to a stored lookup table. BUT I don't agree that an AGI has to create itself when it is obvious humans can't either. Even though Ben believes in emergent intelligence, he has always said that training by humans to at least some level is absolutely necessary. Even though I agree that generalizing is a very desirable quality for an AGI, is this property necessary to creating an AGI? Most people don't generalize all that well in my opinion. David Clark -Original Message- From: William Pearson [mailto:[EMAIL PROTECTED] Sent: March-03-08 8:21 PM To: agi@v2.listbox.com Subject: Re: [agi] Thought experiment on informationally limited systems On 04/03/2008, Mike Tintner [EMAIL PROTECTED] wrote: David: I was specifically referring to your comment ending in BY ITSELF. Jeez, Will, the point of Artificial General Intelligence is that it can start adapting to an unfamiliar situation and domain BY ITSELF. I believe this statement is just plain incorrect. David, I find that extraordinary, but I accept your sincerity. The definition I gave is an essential part of an AGI - if it can't adapt by itself sometimes to unfamiliar situations, (as humans can), and can only act on others' instructions, then it's a narrow AI. I wonder whether anyone else shares your view. There are a number of threads here that need disentangling. All these answers are my opinion only. Is a system that can adapt by itself to unfamiliar situations necessary for AGI? I would answer yes. Is it the only thing an AGI needs to be able to do? No. If I had a system that could build houses out of bricks, stones and straw it would not be an AGI if it could not be taught or learn cryptography. General for me means the ability to learn many different skills, including working on its own. Is generalising a skill logically the first thing that you need to make an AGI? Nope, the means and sufficient architecture to acquire skills and competencies are more useful early on in an agi development. I see generalising, in the way you talk about it, as a skill that can be acquired and improved upon. We certainly can change our ability to do so, through out our life time. If a skill can be changed and/or improved then something in the system must be changed, either data, program or something else. It is the manner and nature of these changes, very low level sub concious stuff (google neuro plasticity for what I am talking about in humans), that I think needs to be worked upon first. Else you are going to create a static and crystalline system. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; 724342 Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
RE: [agi] Thought experiment on informationally limited systems
YOU said teaching an AGI was cheating. YOU want others to talk straight to your points and issues but you make it look like I don't think much of generalizing in an AGI design. The fact is that generalizing is at the heart of my design and is a totally different issue from training being cheating. Ask the question simply: Do the people on this list think that training is necessary for the creation of an AGI and would they call training the AGI cheating? You say training means cheating and to create an AGI you can't do that. I disagree. David Clark -Original Message- From: Mike Tintner [mailto:[EMAIL PROTECTED] Sent: March-03-08 8:48 PM To: agi@v2.listbox.com Subject: Re: [agi] Thought experiment on informationally limited systems Will:Is generalising a skill logically the first thing that you need to make an AGI? Nope, the means and sufficient architecture to acquire skills and competencies are more useful early on in an agi development Ah, you see, that's where I absolutely disagree, and a good part of why I'm hammering on the way I am. I don't think many (anyone?) will agree with David, but many if not everyone will agree with you. Yes, the problem of generalising is the very first thing you tackle, and should shape everything you do - at least once you have moved beyond idle thought to serious engagement. If you're trying to develop a new electric battery, you look for that new chemical first (assuming that's what you reckon you'll need) - you don't start looking at the casing or other aspects of the battery. Anything peripheral you do first may be rendered totally irrelevant later on when you do discover that chemical and a total waste of time. And such, I'm sure, is the case with AGI. That central problem of generalising demands a total new mentality - a sea-change of approach. (You saw an example in my exchange with YKY. I think - in fact, I'm just about totally certain - that generalising demands a system of open- ended concepts like ours. Because he isn't directly concerned with the generalising problem, he wants a closed-ended, unambiguous language - which is in fact only suitable for narrow AI and, I would argue, a waste of time). P.S. It's a bit sad - you started this thread with a generalising problem, now you're backtracking on it. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; 724342 Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
RE: [agi] Thought experiment on informationally limited systems
How intelligent would any human be if it couldn't be taught by other humans? Could a human ever learn to speak by itself? The few times this has happened in real life, the person was permanently disabled and not capable of becoming a normal human being. If humans can't become human without the help of other humans, why should this is a criteria for AGI? David Clark PS I am not suggesting that explicitly programming 100% of an AGI is either doable or desirable but some degree of detailed teaching must be a requirement for all on this list who dream of creating an AGI, no? -Original Message- From: Mike Tintner [mailto:[EMAIL PROTECTED] Sent: March-02-08 5:36 AM To: agi@v2.listbox.com Subject: Re: [agi] Thought experiment on informationally limited systems Jeez, Will, the point of Artificial General Intelligence is that it can start adapting to an unfamiliar situation and domain BY ITSELF. And your FIRST and only response to the problem you set was to say: I'll get someone to tell it what to do. IOW you simply avoided the problem and thought only of cheating. What a solution, or merest idea for a solution, must do is tell me how that intelligence will start adapting by itself - will generalize from its existing skills to cross over domains. Then, as my answer indicated, it may well have to seek some instructions and advice - especially and almost certainly if it wants to acquire a whole new major skill, as we do, by taking courses etc. But a general intelligence should be able to adapt to some unfamiliar situations entirely by itself - like perhaps your submersible situation. No guarantee that it will succeed in any given situation, (as there isn't with us), but you should be able to demonstrate its power to adapt sometimes. In a sense, you should be appalled with yourself that you didn't try to tackle the problem - to produce a cross-over idea. But since literally no one else in the field of AGI has the slightest cross-over idea - i.e. is actually tackling the problem of AGI, - and the whole culture is one of avoiding the problem, it's to be expected. (You disagree - show me one, just one, cross-over idea anywhere. Everyone will give you a v. detailed,impressive timetable for how long it'll take them to produce such an idea, they just will never produce one. Frankly, they're too scared). Mike Tintner [EMAIL PROTECTED] wrote: You must first define its existing skills, then define the new challenge with some degree of precision - then explain the principles by which it will extend its skills. It's those principles of extension/generalization that are the be-all and end-all, (and NOT btw, as you suggest, any helpful info that the robot will receive - that,sir, is cheating - it has to work these things out for itself - although perhaps it could *ask* for info). Why is that cheating? Would you never give instructions to a child about what to do? Taking instuctions is something that all intelligences need to be able to do, but it should be attempted to be minimised. I'm not saying it should take instructions unquestioningly either, ideally it should figure out whether the instructions you give are any use for it. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; 724342 Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
Yes, an AGI will have to be able to do narrow AI. What you are doing here - and everyone is doing over and over and over - is saying: Yes, I know there's a hard part to AGI, but can I please concentrate on the easy parts - the narrow AI parts - first? If I give you a problem, I don't want to know whether you can take dictation and spell, I just want to know whether you can solve the problem - and not make excuses, or create distractions. It's simple - do you have any ideas about the problem of AGI - ideas for generalizing skills (see below) - cross-over ideas - or not? David: How intelligent would any human be if it couldn't be taught by other humans? Could a human ever learn to speak by itself? The few times this has happened in real life, the person was permanently disabled and not capable of becoming a normal human being. If humans can't become human without the help of other humans, why should this is a criteria for AGI? David Clark PS I am not suggesting that explicitly programming 100% of an AGI is either doable or desirable but some degree of detailed teaching must be a requirement for all on this list who dream of creating an AGI, no? -Original Message- From: Mike Tintner [mailto:[EMAIL PROTECTED] Sent: March-02-08 5:36 AM To: agi@v2.listbox.com Subject: Re: [agi] Thought experiment on informationally limited systems Jeez, Will, the point of Artificial General Intelligence is that it can start adapting to an unfamiliar situation and domain BY ITSELF. And your FIRST and only response to the problem you set was to say: I'll get someone to tell it what to do. IOW you simply avoided the problem and thought only of cheating. What a solution, or merest idea for a solution, must do is tell me how that intelligence will start adapting by itself - will generalize from its existing skills to cross over domains. Then, as my answer indicated, it may well have to seek some instructions and advice - especially and almost certainly if it wants to acquire a whole new major skill, as we do, by taking courses etc. But a general intelligence should be able to adapt to some unfamiliar situations entirely by itself - like perhaps your submersible situation. No guarantee that it will succeed in any given situation, (as there isn't with us), but you should be able to demonstrate its power to adapt sometimes. In a sense, you should be appalled with yourself that you didn't try to tackle the problem - to produce a cross-over idea. But since literally no one else in the field of AGI has the slightest cross-over idea - i.e. is actually tackling the problem of AGI, - and the whole culture is one of avoiding the problem, it's to be expected. (You disagree - show me one, just one, cross-over idea anywhere. Everyone will give you a v. detailed,impressive timetable for how long it'll take them to produce such an idea, they just will never produce one. Frankly, they're too scared). Mike Tintner [EMAIL PROTECTED] wrote: You must first define its existing skills, then define the new challenge with some degree of precision - then explain the principles by which it will extend its skills. It's those principles of extension/generalization that are the be-all and end-all, (and NOT btw, as you suggest, any helpful info that the robot will receive - that,sir, is cheating - it has to work these things out for itself - although perhaps it could *ask* for info). Why is that cheating? Would you never give instructions to a child about what to do? Taking instuctions is something that all intelligences need to be able to do, but it should be attempted to be minimised. I'm not saying it should take instructions unquestioningly either, ideally it should figure out whether the instructions you give are any use for it. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; 724342 Powered by Listbox: http://www.listbox.com --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.5.516 / Virus Database: 269.21.3/1308 - Release Date: 3/3/2008 10:01 AM --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
Will:Is generalising a skill logically the first thing that you need to make an AGI? Nope, the means and sufficient architecture to acquire skills and competencies are more useful early on in an agi development Ah, you see, that's where I absolutely disagree, and a good part of why I'm hammering on the way I am. I don't think many (anyone?) will agree with David, but many if not everyone will agree with you. Yes, the problem of generalising is the very first thing you tackle, and should shape everything you do - at least once you have moved beyond idle thought to serious engagement. If you're trying to develop a new electric battery, you look for that new chemical first (assuming that's what you reckon you'll need) - you don't start looking at the casing or other aspects of the battery. Anything peripheral you do first may be rendered totally irrelevant later on when you do discover that chemical and a total waste of time. And such, I'm sure, is the case with AGI. That central problem of generalising demands a total new mentality - a sea-change of approach. (You saw an example in my exchange with YKY. I think - in fact, I'm just about totally certain - that generalising demands a system of open-ended concepts like ours. Because he isn't directly concerned with the generalising problem, he wants a closed-ended, unambiguous language - which is in fact only suitable for narrow AI and, I would argue, a waste of time). P.S. It's a bit sad - you started this thread with a generalising problem, now you're backtracking on it. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
On 28/02/2008, YKY (Yan King Yin) [EMAIL PROTECTED] wrote: On 2/28/08, William Pearson [EMAIL PROTECTED] wrote: Note I want something different than computational universality. E.g. Von Neumann architectures are generally programmable, Harvard architectures aren't. As they can't be reprogrammed at run time. It seems that you want to build the AGI from the programming level. This is in contrast to John MacCarthy's declarative paradigm. Your approach offers more flexibility (perhaps maximum flexibility), but may not make AGI easier to build. Learning, in your case, is a matter of algorithmic learning. It may be harder / less efficient than logic-based learning. Algorithmic learning is hard. But just because the system is based upon programs as its lowest level representation, does not mean that all learning is going to be algorithmic learning. It is possible to have programs that learn in any fashion within the system. If it makes sense in the system, you could have a logic based learning program. Just that it will be in competition with other learners to see which is the most useful for the system. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
On 28/02/2008, Mike Tintner [EMAIL PROTECTED] wrote: You must first define its existing skills, then define the new challenge with some degree of precision - then explain the principles by which it will extend its skills. It's those principles of extension/generalization that are the be-all and end-all, (and NOT btw, as you suggest, any helpful info that the robot will receive - that,sir, is cheating - it has to work these things out for itself - although perhaps it could *ask* for info). Why is that cheating? Would you never give instructions to a child about what to do? Taking instuctions is something that all intelligences need to be able to do, but it should be attempted to be minimised. I'm not saying it should take instructions unquestioningly either, ideally it should figure out whether the instructions you give are any use for it. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
Jeez, Will, the point of Artificial General Intelligence is that it can start adapting to an unfamiliar situation and domain BY ITSELF. And your FIRST and only response to the problem you set was to say: I'll get someone to tell it what to do. IOW you simply avoided the problem and thought only of cheating. What a solution, or merest idea for a solution, must do is tell me how that intelligence will start adapting by itself - will generalize from its existing skills to cross over domains. Then, as my answer indicated, it may well have to seek some instructions and advice - especially and almost certainly if it wants to acquire a whole new major skill, as we do, by taking courses etc. But a general intelligence should be able to adapt to some unfamiliar situations entirely by itself - like perhaps your submersible situation. No guarantee that it will succeed in any given situation, (as there isn't with us), but you should be able to demonstrate its power to adapt sometimes. In a sense, you should be appalled with yourself that you didn't try to tackle the problem - to produce a cross-over idea. But since literally no one else in the field of AGI has the slightest cross-over idea - i.e. is actually tackling the problem of AGI, - and the whole culture is one of avoiding the problem, it's to be expected. (You disagree - show me one, just one, cross-over idea anywhere. Everyone will give you a v. detailed,impressive timetable for how long it'll take them to produce such an idea, they just will never produce one. Frankly, they're too scared). Mike Tintner [EMAIL PROTECTED] wrote: You must first define its existing skills, then define the new challenge with some degree of precision - then explain the principles by which it will extend its skills. It's those principles of extension/generalization that are the be-all and end-all, (and NOT btw, as you suggest, any helpful info that the robot will receive - that,sir, is cheating - it has to work these things out for itself - although perhaps it could *ask* for info). Why is that cheating? Would you never give instructions to a child about what to do? Taking instuctions is something that all intelligences need to be able to do, but it should be attempted to be minimised. I'm not saying it should take instructions unquestioningly either, ideally it should figure out whether the instructions you give are any use for it. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
One thing I would expect from an AGI is that it least it would be able to Google for something that might talk about how to do whatever it needs and to have available library references on the subject. Being able to follow and interpret written instructions takes a lot of intelligence in itself. And a lot of times there are important conventions about how to do certain things and it is a bad idea to just do things completely your own way. Of course, we do expect an intelligent agent to often be able to figure things out on its own. But you have to remember that when you allow for this, there will sometimes be mistakes. It is a necessary consequence of trying something new that it won't always work. And I am afraid that people have an unrealistic expectation that the AGI will be able to do something new without ever getting it wrong. I'm expecting there will be a lot of pain and disappointment when it doesn't work this way. Because it can't work that way. An AGI working in unknown territory will have to make mistakes. Andi Quoting Mike Tintner [EMAIL PROTECTED]: Jeez, Will, the point of Artificial General Intelligence is that it can start adapting to an unfamiliar situation and domain BY ITSELF. And your FIRST and only response to the problem you set was to say: I'll get someone to tell it what to do. IOW you simply avoided the problem and thought only of cheating. What a solution, or merest idea for a solution, must do is tell me how that intelligence will start adapting by itself - will generalize from its existing skills to cross over domains. Then, as my answer indicated, it may well have to seek some instructions and advice - especially and almost certainly if it wants to acquire a whole new major skill, as we do, by taking courses etc. But a general intelligence should be able to adapt to some unfamiliar situations entirely by itself - like perhaps your submersible situation. No guarantee that it will succeed in any given situation, (as there isn't with us), but you should be able to demonstrate its power to adapt sometimes. In a sense, you should be appalled with yourself that you didn't try to tackle the problem - to produce a cross-over idea. But since literally no one else in the field of AGI has the slightest cross-over idea - i.e. is actually tackling the problem of AGI, - and the whole culture is one of avoiding the problem, it's to be expected. (You disagree - show me one, just one, cross-over idea anywhere. Everyone will give you a v. detailed,impressive timetable for how long it'll take them to produce such an idea, they just will never produce one. Frankly, they're too scared). Mike Tintner [EMAIL PROTECTED] wrote: You must first define its existing skills, then define the new challenge with some degree of precision - then explain the principles by which it will extend its skills. It's those principles of extension/generalization that are the be-all and end-all, (and NOT btw, as you suggest, any helpful info that the robot will receive - that,sir, is cheating - it has to work these things out for itself - although perhaps it could *ask* for info). Why is that cheating? Would you never give instructions to a child about what to do? Taking instuctions is something that all intelligences need to be able to do, but it should be attempted to be minimised. I'm not saying it should take instructions unquestioningly either, ideally it should figure out whether the instructions you give are any use for it. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
Yes of course an AGI will make mistakes - and sometimes fail - in adapting. I say that v. explicitly. But your other point also skirts the problem - which is that the AGI must first identify what it needs to adapt to, before it can start googling/asking for advice. I think we need better to focus the problem. I gave I think a good example. How is a system going to build a wall if the only materials it knows to build a wall with - bricks - are unavailable? You might care to focus the submersible problem. Let's say it is that the submersible finds it cannot rise - it just won't go upwards. I'm not v. mechanical so you guys can perhaps flesh it out. Something is preventing it from rising, but all the obvious things are functioning OK - define what it knows, A-X, define what the mysterious problem is, Z, and then you have a true AGI problem - how does it generalize fromA-X or any other knowledge to Z (a v. different domain)? Z, for example, might be a squid (unless it already knows about squids). A good deal of imagination has to go into just defining AGI problems - you have to spend a good deal of time on it. Andi: One thing I would expect from an AGI is that it least it would be able to Google for something that might talk about how to do whatever it needs and to have available library references on the subject. Being able to follow and interpret written instructions takes a lot of intelligence in itself. And a lot of times there are important conventions about how to do certain things and it is a bad idea to just do things completely your own way. Of course, we do expect an intelligent agent to often be able to figure things out on its own. But you have to remember that when you allow for this, there will sometimes be mistakes. It is a necessary consequence of trying something new that it won't always work. And I am afraid that people have an unrealistic expectation that the AGI will be able to do something new without ever getting it wrong. I'm expecting there will be a lot of pain and disappointment when it doesn't work this way. Because it can't work that way. An AGI working in unknown territory will have to make mistakes. Andi Quoting Mike Tintner [EMAIL PROTECTED]: Jeez, Will, the point of Artificial General Intelligence is that it can start adapting to an unfamiliar situation and domain BY ITSELF. And your FIRST and only response to the problem you set was to say: I'll get someone to tell it what to do. IOW you simply avoided the problem and thought only of cheating. What a solution, or merest idea for a solution, must do is tell me how that intelligence will start adapting by itself - will generalize from its existing skills to cross over domains. Then, as my answer indicated, it may well have to seek some instructions and advice - especially and almost certainly if it wants to acquire a whole new major skill, as we do, by taking courses etc. But a general intelligence should be able to adapt to some unfamiliar situations entirely by itself - like perhaps your submersible situation. No guarantee that it will succeed in any given situation, (as there isn't with us), but you should be able to demonstrate its power to adapt sometimes. In a sense, you should be appalled with yourself that you didn't try to tackle the problem - to produce a cross-over idea. But since literally no one else in the field of AGI has the slightest cross-over idea - i.e. is actually tackling the problem of AGI, - and the whole culture is one of avoiding the problem, it's to be expected. (You disagree - show me one, just one, cross-over idea anywhere. Everyone will give you a v. detailed,impressive timetable for how long it'll take them to produce such an idea, they just will never produce one. Frankly, they're too scared). Mike Tintner [EMAIL PROTECTED] wrote: You must first define its existing skills, then define the new challenge with some degree of precision - then explain the principles by which it will extend its skills. It's those principles of extension/generalization that are the be-all and end-all, (and NOT btw, as you suggest, any helpful info that the robot will receive - that,sir, is cheating - it has to work these things out for itself - although perhaps it could *ask* for info). Why is that cheating? Would you never give instructions to a child about what to do? Taking instuctions is something that all intelligences need to be able to do, but it should be attempted to be minimised. I'm not saying it should take instructions unquestioningly either, ideally it should figure out whether the instructions you give are any use for it. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com -- No virus found in this incoming message.
Re: [agi] Thought experiment on informationally limited systems
On 02/03/2008, Mike Tintner [EMAIL PROTECTED] wrote: Jeez, Will, the point of Artificial General Intelligence is that it can start adapting to an unfamiliar situation and domain BY ITSELF. And your FIRST and only response to the problem you set was to say: I'll get someone to tell it what to do. Nothing we ever do is by ourselves, entirely, we have a wealth of examples to draw from that we have acquired from family/friends and teachers. The situation I described was like throwing a baby into a completely unfamiliar problem, without the wealth of experience we have built up over the years, so some hand holding is to be expected. Also I'm not planning to have a full AI made any time soon, I'm merely laying the ground work, for many other people to work upon. I may get animal level adaptivity/intelligence myself, it depends how quickly I can build the first layer and the tools I need for the next. This is also why I concentrate on the most flexible system possible, I do not wish to constrain the system to do any more than needs be done to achieve my current goal. This goal is to add a way of selecting between the programs within a computer system dependent upon what the system needs to do. It is more fundamental than your cross-over idea, in that it is a lower level phenomenon, but not in the sense it is more important for acting intelligently. IOW you simply avoided the problem and thought only of cheating. What a solution, or merest idea for a solution, must do is tell me how that intelligence will start adapting by itself - will generalize from its existing skills to cross over domains. I'm not building the solution, merely a framework which I think will enable people to build the solution. I think this needs to be done first, in essence I am trying to deal with the develop and acquire skills problem. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
Although top down should continue being researched tried, the complexity is still monumental. We KNOW that bottom up delivers AGI, and Turing's view was that heuristics are enough to build it. That is only doable at mass speeds assumed possible in eg quantum computing. eldras - Original Message - From: William Pearson [EMAIL PROTECTED] To: agi@v2.listbox.com Subject: Re: [agi] Thought experiment on informationally limited systems Date: Sun, 2 Mar 2008 23:04:27 + On 02/03/2008, Mike Tintner [EMAIL PROTECTED] wrote: Jeez, Will, the point of Artificial General Intelligence is that it can start adapting to an unfamiliar situation and domain BY ITSELF. And your FIRST and only response to the problem you set was to say: I'll get someone to tell it what to do. Nothing we ever do is by ourselves, entirely, we have a wealth of examples to draw from that we have acquired from family/friends and teachers. The situation I described was like throwing a baby into a completely unfamiliar problem, without the wealth of experience we have built up over the years, so some hand holding is to be expected. Also I'm not planning to have a full AI made any time soon, I'm merely laying the ground work, for many other people to work upon. I may get animal level adaptivity/intelligence myself, it depends how quickly I can build the first layer and the tools I need for the next. This is also why I concentrate on the most flexible system possible, I do not wish to constrain the system to do any more than needs be done to achieve my current goal. This goal is to add a way of selecting between the programs within a computer system dependent upon what the system needs to do. It is more fundamental than your cross-over idea, in that it is a lower level phenomenon, but not in the sense it is more important for acting intelligently. IOW you simply avoided the problem and thought only of cheating. What a solution, or merest idea for a solution, must do is tell me how that intelligence will start adapting by itself - will generalize from its existing skills to cross over domains. I'm not building the solution, merely a framework which I think will enable people to build the solution. I think this needs to be done first, in essence I am trying to deal with the develop and acquire skills problem. Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com -- Want an e-mail address like mine? Get a free e-mail account today at www.mail.com! --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
On 2/28/08, William Pearson [EMAIL PROTECTED] wrote: I'm going to try and elucidate my approach to building an intelligent system, in a round about fashion. This is the problem I am trying to solve. Imagine you are designing a computer system to solve an unknown problem, and you have these constraints A) Limited space to put general information about the world B) Communication with the system after it has been deployed. The less the better. C) We shall also assume limited processing ability etc The goal is to create a system that can solve the tasks as quickly as possible with the least interference from the outside. I'd like people to write a brief sketch of your solution to this sort of problem down. Is it different from your AGI designs, if so why? Space/time-optimality is not my top concern. I'm focused on building an AGI that *works*, within reasonable space/time. If you add these contraints, you're making the AGI problem harder than it already is. Ditto for the amount of user interaction. Why make it harder? System Sketch? - It would have to be generally programmable, I would want to be able to send it arbitrary programs after it had been created, so I could send it a program to decrypt things or control things. It would also need to able to generate it's own programming and select between the different programs in order to minimise my need to program it. It is not different to my AGI design, unsurprisingly. Generally programmable, yes. But that's very broad. Many systems have this property. Even system with only a declarative KB can re-program itself by modifying the KB. YKY --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
I guess the first thing you would need for an Unknown Problem Solver would be some way to determine usefulness. To be able to achieve some goal the system may need measures of usefulness which span intermediate stages towards the goal, or which are stacked in a series. If the system has no idea of usefulness and no explicit goals then probably the best it can do is become an error corrector - i.e. look for regular patterns of activity, then find anomalies and try to take actions which correct those anomalies and restore the expected pattern. On 28/02/2008, William Pearson [EMAIL PROTECTED] wrote: I'm going to try and elucidate my approach to building an intelligent system, in a round about fashion. This is the problem I am trying to solve. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
On 28/02/2008, YKY (Yan King Yin) [EMAIL PROTECTED] wrote: On 2/28/08, William Pearson [EMAIL PROTECTED] wrote: I'm going to try and elucidate my approach to building an intelligent system, in a round about fashion. This is the problem I am trying to solve. Imagine you are designing a computer system to solve an unknown problem, and you have these constraints A) Limited space to put general information about the world B) Communication with the system after it has been deployed. The less the better. C) We shall also assume limited processing ability etc The goal is to create a system that can solve the tasks as quickly as possible with the least interference from the outside. I'd like people to write a brief sketch of your solution to this sort of problem down. Is it different from your AGI designs, if so why? Space/time-optimality is not my top concern. I'm focused on building an AGI that *works*, within reasonable space/time. If you add these contraints, you're making the AGI problem harder than it already is. Ditto for the amount of user interaction. Why make it harder? I'm not looking for optimality, just that better is important. I don't want to have to hold the hand of my system teaching it laboriously, so the less information I have to feed it the better. Why ignore the problem and make the job of teaching it harder? Also we have limited space and time in the real world System Sketch? - It would have to be generally programmable, I would want to be able to send it arbitrary programs after it had been created, so I could send it a program to decrypt things or control things. It would also need to able to generate it's own programming and select between the different programs in order to minimise my need to program it. It is not different to my AGI design, unsurprisingly. Generally programmable, yes. But that's very broad. Many systems have this property. Note I want something different than computational universality. E.g. Von Neumann architectures are generally programmable, Harvard architectures aren't. As they can't be reprogrammed at run time. http://en.wikipedia.org/wiki/Harvard_architecture Even system with only a declarative KB can re-program itself by modifying the KB. So a program could get in and remove all the items from the KB? You can have viruses etc inside the KB? Will Pearson --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
On 2/28/08, William Pearson [EMAIL PROTECTED] wrote: Note I want something different than computational universality. E.g. Von Neumann architectures are generally programmable, Harvard architectures aren't. As they can't be reprogrammed at run time. It seems that you want to build the AGI from the programming level. This is in contrast to John MacCarthy's declarative paradigm. Your approach offers more flexibility (perhaps maximum flexibility), but may not make AGI easier to build. Learning, in your case, is a matter of algorithmic learning. It may be harder / less efficient than logic-based learning. YKY --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
WP: I'm going to try and elucidate my approach to building an intelligent system, in a round about fashion. This is the problem I am trying to solve. Marks for at least trying to identify an AGI problem. I can't recall anyone else doing so - which, to repeat, I think is appalling. But I don't think you're doing it adequately, with your example of a submersible. Essentially you're providing a variation on what I've already mentioned - the ICRA Robot Challenge. http://icra.wustl.edu/ Their lunar mission camp situation - how would you program a robot to deal with any unexpected emergency that could arise in a limited camp area - any malfunctioning equipment, for example - does strike me as a good AGI problem. Your robot will have certain skills: how will it adapt those skills to meet new problems for which they are useful but, overall inadequate? You must first define its existing skills, then define the new challenge with some degree of precision - then explain the principles by which it will extend its skills. It's those principles of extension/generalization that are the be-all and end-all, (and NOT btw, as you suggest, any helpful info that the robot will receive - that,sir, is cheating - it has to work these things out for itself - although perhaps it could *ask* for info). Anyway, nice to see someone talking about the central problem of AGI at last. --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
On Thu, Feb 28, 2008 at 3:20 PM, William Pearson [EMAIL PROTECTED] wrote: On 28/02/2008, YKY (Yan King Yin) [EMAIL PROTECTED] wrote: Generally programmable, yes. But that's very broad. Many systems have this property. Note I want something different than computational universality. E.g. Von Neumann architectures are generally programmable, Harvard architectures aren't. As they can't be reprogrammed at run time. http://en.wikipedia.org/wiki/Harvard_architecture I agree with YKY, it's not a very useful specification. Turing machine is not necessarily a way either. I think that ability to learn structure-less production rules is sufficient. These allow system to implement finite state machines internally, and these state machines operate on data streams consisting of external I/O and states of other state machines. This organizational principle follows naturally from blackboard-like system where most of the facts on the blackboard are labels given to statistical regularities detected in previous moments. -- Vladimir Nesov [EMAIL PROTECTED] --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com
Re: [agi] Thought experiment on informationally limited systems
--- William Pearson [EMAIL PROTECTED] wrote: I'm going to try and elucidate my approach to building an intelligent system, in a round about fashion. This is the problem I am trying to solve. Imagine you are designing a computer system to solve an unknown problem, and you have these constraints A) Limited space to put general information about the world B) Communication with the system after it has been deployed. The less the better. C) We shall also assume limited processing ability etc The goal is to create a system that can solve the tasks as quickly as possible with the least interference from the outside. I'd like people to write a brief sketch of your solution to this sort of problem down. Is it different from your AGI designs, if so why? The general problem is not computable, like AIXI or compression. So I have to make a guess as to what unknown problems the AGI would be asked to solve. My guess would be problems that have economic value. So I would look at the kind of tasks that people are being paid to solve, and design the AGI to solve the same kind of problems. I would tell the AGI what I want it to accomplish, and it would research the internet to find a good solution. To do that the AGI will first need natural language capability, followed by vision, hearing, and mobility depending on the range of tasks. It will need vastly more computing power than the average PC in any case. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: http://www.listbox.com/member/archive/303/=now RSS Feed: http://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: http://www.listbox.com/member/?member_id=8660244id_secret=95818715-a78a9b Powered by Listbox: http://www.listbox.com