Re: [agi] Breaking AIXI-tl
Eliezer S. Yudkowsky asked Ben Goertzel: Do you have a non-intuitive mental simulation mode? LOL --#:^D It *is* a valid question, Eliezer, but it makes me laugh. Michael Roy Ames [Who currently estimates his *non-intuitive mental simulation mode* to contain about 3 iterations of 5 variables each - 8 variables each on a good day. Each variable can link to a concept (either complex or simple)... and if that sounds to you like something that a trashed-out Commodore 64 could emulate, then you have some idea how he feels being stuck at his current level of non-intuitive intelligence.] --- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/?[EMAIL PROTECTED]
Re: [agi] AI Morality -- a hopeless quest
Arthur T. Murray wrote: [snippage] why should we creators of Strong AI have to take any more precautions with our Moravecian Mind Children than human parents do with their human babies? Here are three reasons I can think of, Arthur: 1) Because we know in advance that 'Strong AI', as you put it, will be very much smarter and very much more capable than we are - that is not true in the human scenario. 2) If we don't get AI morality right the first time (or very close to it), its game over for the human race. 3) Attempting to develop 'Strong AI' without spending time getting the morality-bit correct, may cause a governmental agency to squash you like a bug. And I didn't even have to think very hard to come up with those... I'm sure there are other reasons. Could you articulate the reasons why you think the 'quest' is hopeless? Michael Roy Ames --- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/?[EMAIL PROTECTED]
Re: [agi] AI Morality -- a hopeless quest
Brad Wyble wrote: I don't think any human alive has the moral and ethical underpinnings to allow them to resist the corruption of absolute power in the long run. I am exceedingly glad that I do not share your opinion on this. Human altruism *is* possible, and indeed I observe myself possessing a significant measure of it. Anyone doubting thier ability to 'resist corruption' should not IMO be working in AGI, but should be doing some serious introspection/study of thier goals and motivations. (No offence intended, Brad) Michael Roy Ames --- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/?[EMAIL PROTECTED]
Re: [agi] AI Morality -- a hopeless quest
Brad Wyble wrote: Under the ethical code you describe, the AGI would swat them like a bug with no more concern than you swatting a mosquito. I did not describe an ethical code, I described two scenarios about a human (myself) then suggested the non-bug-swatting scenario was possible, analogically, for an AGI. All I'm trying to do is shift the focus for a few moments to our own ethical standards as people. If we were put into the shoes of an AGI, would we behave well towards the inferior species? I presume from the phrase If we were put into the shoes of an AGI that human morality and ethics would come along for the ride. If that is what you meant: then it depends on which human you pick as to what happens. I have observed both altruism and cruelty, obsession and indifference in human behaviour toward other species. It bears some thinking about just exactly what one would do in such a situation... I know I have often thought about it. Philip brings up the point that a community AGI's could possibly self-police. I agree. I don't. Policing is only useful/meaningful within a community of almost equal actors that have very little real power. If the actors are not almost equally powerful then you have the 'human and a bug' scenario. If the actors have a very large amount of power, then a single 'transgression' could wipe us all out before any 'policing action' could be initiated. Nor, would one presume, on an AGI's. They might end up with it anyway. I would not presume that so readily. Taking it as a given that we are discussion a Friendly AGI, I would say that there would be significant utility in obtaining a great deal of power. Not to 'Lord it over the petty humans', but to protect them both internal and external threats. Michael Roy Ames --- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/?[EMAIL PROTECTED]
[agi] Re: Games for AIs
Tony, Thanks for sharing your ideas (sorry for the erroneous naming of shape-world). We seem to agree that the lessons (for an AI) would need to start of very simple, and gradually build up mental tools techniques, by using many different games to build slightly different aspects of cognition. The idea of putting a baby AI in a simulated world where it might learn cognitive skills is appealing. But I suspect that it will take a huge number of iterations for the baby AI to learn the needed lessons in that situation. I think it will be faster to give more constrained and structured learning first, then when the AI is capable of understanding the 'game world' and the 'game rules' and the 'game interface' it could play Video games with the intention of *discovering* how it all works together. This is often what humans find interesting about video games: the discovery aspect. And this would be a valuable new skill to develop: given this World, these Rules and this Interface, discover A, B, C. Where A, B, C could be many different things. Eg: Reach the highest level possible; Stay alive the longest; Rack-up the most points; Rack-up the least points without dying... But this kind of excercise is only going to be useful for mental development once the AI has very significant capabilities. I might almost say that, by the time video-game-playing becomes beneficial to mental development, the AI would be largely self-directed. Hey! Alan! Wanna play Duke Nukem? Okay Michael. But I'm going to win this time. Oh, really? How do you know. Well, its a bit complicated to explain. Why don't I just show you? (gulp) Michael Roy Ames Tony Lofthouse wrote: Michael, You wrote: Tony Lofthouse: I've heard you are working on the shape-world interface. Have you considered what games we might play in it? Ideas? To clarify this point. I am currently developing a 2D input capability for Novamente. It is a very crude form of vision that allows the presentation of (x, y) time series to the system. This should not be confused with the shape-world interface mentioned above. Whilst one may lead to the other shape-world is not the current focus. Having said this I do have a couple of comments relating to AI games. Those of you who have had the opportunity to raise children will no doubt be well aware of the fact that children don't play TLoZ (or contemporary equivalent) until well into their childhood. There are many stages of learning before a child is capable of this level of sophistication. One of the first games that young children play is the categorisation game, i.e. What shape is this?, what colour is this?, how may sides?, etc. I would expect to use the 2D world and Shape-world subsequently for the same purpose. This is followed by the comparison game, i.e. is this big?, is this small?, which is bigger?, etc. Then you have the counting game (sort of obvious). The relationship game, i.e. above, below, inside, outside. There are lots of these type games! Then you move on to the reasoning game, i.e. what comes next?, what is missing?, what is the odd one out?, etc. Now the child is ready to combine learning from these different games and moves on to story telling both listening to them and then telling them. Then there are several more years of honing these key skills whilst increasing the level of world knowledge and social understanding. Finally the child is ready to play TLoZ! So as you can see I think there is a lot to do before you get to play TLoZ with your baby AGI. That is the purpose of 2d World and then Shape-World. T --- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/?[EMAIL PROTECTED]
Re: [agi] introduction
Damien Sullivan wrote: Hi! I joined this list recently, figured I'd say who I am. Well, some of you may know already, from extropians, where I used to post a fair bit :) or from my Vernor Vinge page. But now I'm a first year comp sci/cog sci PhD student at Indiana University, hoping to work on extending Jim Marshall's Metacat in Hofstadter's lab. Nothing much has really happened beyond hope and a few meetings and taking his group theory class. I've been reading Eliezer's _Levels_ pages, and having Andy Clark's _Being There_ around, but mostly my life has been classes. Mostly the OS class, actually. Sigh. -xx- Damien X-) --- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/?[EMAIL PROTECTED] Damien, Hi. I'm am quite interested in Jim's Metacat also. It's on my To-Do list to get it running under linux... but the way my workload is going I think Jim will get his planned re-write done first. :)It would be interesting to hear about what new directions Metacat is going in. Welcome to the list. Michael Roy Ames --- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/?[EMAIL PROTECTED]