Re: [agi] What would motivate you to put work into an AGI project?
Interesting question you raise there, Matt (vs :) YKY How many of us would be prepared to work FULL-TIME on AGI: (0) If a department of defense/military organisation paid you develop a secret AGI for national defense/intelligence purpose? (1) If a Microsoft, Google, Sun or IBM came along and hired you full-time to work on either (1a) Open-Source; or (1b) Proprietary AGI? (2) A more 'friendly' research group came along (e.g. University, government agency) to pay you fulltime (2a) on *their* design/architecture or (2b) on YOUR design but having to share your findings with the larger community (shared credit)? (3) If you had sufficient funds of your own? Re (3) I have often wondered how much time one could really spend continuously on working on AGI - refer to the Princeton Instititue of Advanced Studies where established geniuses (such as Einstein) were/are paid to devote fulltime efforts to thinking but actually not many earthshaking ideas have come out of it. Don't we need a lot of 'time wasted' on trivia such as a real job, leaking plumbing and family in order to have these 1 or 2 hours of creative thinking/work each day? Would it help to have consolidated 8 hour or longer blocks each day? Do people like Ben, Leitz and Peter (Voss) really have so much time to think creatively/design or is my suspicion right that a lot of their (your :) time is spent on fundraising, PR, communication, management? The grass always seems greener on the other side... Jean-Paul Matt Mahoney [EMAIL PROTECTED] 2007/05/07 03:47:28 I think we should not go FOSS just because we arn't confident of ourselves, or to try to avoid competition. We love our work and should go the extra miles to make it profitable. Those who're not interested in business matters can leave that to somebody else in the group. The problem with closed source is you have to pay your employees. Personally, I am not interested in making a lot of money. I already make enough to buy what I want. It is more important to have free time to pursue my interests. AGI, especially language, is one of my interests. But I don't want to build something aimlessly like Cyc. I would like to see an application, a goal in which progress can be measured. I currently use text compression for this purpose. Do you have a better idea? - 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 Advantages of a Conscious Mind
On 5/6/07, Mike Tintner [EMAIL PROTECTED] wrote: Pei, I assumed your system is determinisitc from your posts, not your papers. So I'm still really, genuinely confused by your position. You didn't actually answer my question (unless I've missed something in all these posts) re how your system could have a choice and yet not be arbitrary at all. Mike, Email discussion is a supplement, not a replacement, of literature reading. If you are really serious about this discussion, you have no choice but to read much more than you did. For a topic as complicated and difficult as AGI, one shouldn't expect to resolve all issues by email. For example, your following comments show me that you don't share even the minimum common knowledge and terminology with people studying decision making. By your definition, free will have to remain magical, since any successful explanation will turn the decision making process into deterministic. As several people have pointed out, you can believe whatever you want, but to carry out a fruitful discussion, you have to follow the common convention of communication. Even if you really have revolutionary ideas, you need to express them in acceptable ways. Otherwise it is simply a waste of time, both yours and other people's. I have tried my best in answering your questions. Pei Listen, you can define your system any which way you like. Why not do it simply and directly? A free system can decide at a given point, either of two or multiple ways, - in my example, to Buy, Sell or Hold. A deterministic system at that same point, will have only one option. It will have, say, to decide to Sell. Which is your system? (Philosophers may argue till the end of time about what is/ isn't compatibilist, incompatibilisit, etc etc but they won't define free and determined decisionmaking any differently). - 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 Advantages of a Conscious Mind
J. Storrs Hall, PhD. writes: NVIDIA claims half a teraflop for the 8800 gtx. You need an embarassingly parallel problem, tho. That claim is slightly bogus (I think they are figuring in some graphics-specific feature which would rarely if ever be used by general purpose algorithms [texture interpolation?]). The actual numbers are pretty easy to compute, I think: 128 stream processors x 1.35GHz clock = 173 billion instructions per second. Since one of the instructions is a Multiply-Add, couble that to a 346 gigaflops peak. Counting a multiply-add instruction as a two-flop is probably okay because so many algorithms can make use of it (matrix operations, convolutions, etc). As is usually the case, memory bandwidth is a bigger issue. Access to the 768mb card memory has a bandwidth of about 80 gb/sec, which means that to keep the processors busy, one needs a compute intensity of about 8 instructions for each load of a 4-byte float. This is the primary reason that most computations don't hit the peak numbers -- for example, multiplying large matrices using their libraries can hit 100 gflops but I don't think they have improved it beyond that. Convolution could do somewhat better I think, partially because the kernel can be saved in on-chip memory. Latency complicates the programming -- when data needs to be fetched from card memory, figure 200 cycles for it to get delivered. Other threads can get useful work done during that time, but for that to happen there has to be a huge number of threads (thus the embarassingly parallel comment you made). The biggest limitation, of course, is the bandwidth between the main system memory and the card memory over the pci express bus, where one is lucky to get 2 gb/sec. Sorry if this doesn't seem too much like cognitive science; I don't think it's completely off-topic though to talk about the computational resources available to AGI research. - 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
Time Enough For Work [WAS Re: [agi] What would motivate you to put work into an AGI project?]
Myself, I think that the number of hours working alone might only need to be a small number (3-4). But what I value most is hours brainstorming with others who are of like mind and similar level of knowledge. That is a gold-dust situation. I have been watching From The Earth To The Moon recently. Oh to be part of a 100,000-strong community working on one noble project! Richard Loosemore. Jean-Paul Van Belle wrote: Interesting question you raise there, Matt (vs :) YKY How many of us would be prepared to work FULL-TIME on AGI: (0) If a department of defense/military organisation paid you develop a secret AGI for national defense/intelligence purpose? (1) If a Microsoft, Google, Sun or IBM came along and hired you full-time to work on either (1a) Open-Source; or (1b) Proprietary AGI? (2) A more 'friendly' research group came along (e.g. University, government agency) to pay you fulltime (2a) on *their* design/architecture or (2b) on YOUR design but having to share your findings with the larger community (shared credit)? (3) If you had sufficient funds of your own? Re (3) I have often wondered how much time one could really spend continuously on working on AGI - refer to the Princeton Instititue of Advanced Studies where established geniuses (such as Einstein) were/are paid to devote fulltime efforts to thinking but actually not many earthshaking ideas have come out of it. Don't we need a lot of 'time wasted' on trivia such as a real job, leaking plumbing and family in order to have these 1 or 2 hours of creative thinking/work each day? Would it help to have consolidated 8 hour or longer blocks each day? Do people like Ben, Leitz and Peter (Voss) really have so much time to think creatively/design or is my suspicion right that a lot of their (your :) time is spent on fundraising, PR, communication, management? The grass always seems greener on the other side... Jean-Paul - 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: Time Enough For Work [WAS Re: [agi] What would motivate you to put work into an AGI project?]
I've found that if I can do 16 to 24 hours on, and then sleep, and then another 16 to 24 hours on as long as the body can keep up I can reach really high thresholds of productivity with ephemeral visions of deep, comprehensive insight. In the past I've done self-directed data compression RD, written video games, telephony switching software in this way on up to 2 year stints. 8 hours is just warming up. And the concept of the Long Day Society I've kicked about for years where the day I think is too short at 24 hours. Stretching the wake sleep cycle could help us live longer But the software I'm working on (WKG - Web Knowledge Gatherer) is pre-AGI and at some point will need to grow a brain :) so reading all these discussions and interactions especially among the more astute and learned individuals on this email list is very informative gives perspective on some of the RD that is going on. And any posts and references on good learning material and books are helpful. I've just started reading The Symbolic Species by Terrence W. Deacon which has sat on my shelf for years :) and is a little outdated I suppose but has some good information and provides some examples examining the human brain but ... brain and computer software s different and the brain is such a conglomerated mish-mash of evolutionary cognitive appendages! It's almost like OK need to start from scratch when building AGI like when software becomes fragile, rigid, brittle, and rots (as Agile design tries to avoid). If the brain could be decompiled, which I'm sure we are getting closer, how much of it would really be useful for AGI? Would the source code be too messy and spaghetti like? Are there any algorithms and data structures that haven't been discovered in mathematics? And modeling AGI after brain... maybe loosely. Do we model machines after human body design, some yes but others not, say an army tank is in some ways like a human body but it more accommodates humans (and disaccommodates) than is modeled after. The decompiled brain source code would have some really amazing undiscovered stuff in there. Maybe it couldn't be represented with conventional programming languages. I wonder... parts of it would be immensely sophisticated yes that's a de facto assumption :). And it could be decompiled at many levels. I suppose a functional level, macroscopic decompiler could be made now there are probably many in existence... John -Original Message- From: Richard Loosemore [mailto:[EMAIL PROTECTED] Sent: Monday, May 07, 2007 9:58 AM To: agi@v2.listbox.com Subject: Time Enough For Work [WAS Re: [agi] What would motivate you to put work into an AGI project?] Myself, I think that the number of hours working alone might only need to be a small number (3-4). But what I value most is hours brainstorming with others who are of like mind and similar level of knowledge. That is a gold-dust situation. I have been watching From The Earth To The Moon recently. Oh to be part of a 100,000-strong community working on one noble project! Richard Loosemore. Jean-Paul Van Belle wrote: Interesting question you raise there, Matt (vs :) YKY How many of us would be prepared to work FULL-TIME on AGI: (0) If a department of defense/military organisation paid you develop a secret AGI for national defense/intelligence purpose? (1) If a Microsoft, Google, Sun or IBM came along and hired you full-time to work on either (1a) Open-Source; or (1b) Proprietary AGI? (2) A more 'friendly' research group came along (e.g. University, government agency) to pay you fulltime (2a) on *their* design/architecture or (2b) on YOUR design but having to share your findings with the larger community (shared credit)? (3) If you had sufficient funds of your own? Re (3) I have often wondered how much time one could really spend continuously on working on AGI - refer to the Princeton Instititue of Advanced Studies where established geniuses (such as Einstein) were/are paid to devote fulltime efforts to thinking but actually not many earthshaking ideas have come out of it. Don't we need a lot of 'time wasted' on trivia such as a real job, leaking plumbing and family in order to have these 1 or 2 hours of creative thinking/work each day? Would it help to have consolidated 8 hour or longer blocks each day? Do people like Ben, Leitz and Peter (Voss) really have so much time to think creatively/design or is my suspicion right that a lot of their (your :) time is spent on fundraising, PR, communication, management? The grass always seems greener on the other side... Jean-Paul - 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: Time Enough For Work [WAS Re: [agi] What would motivate you to put work into an AGI project?]
Re (3) I have often wondered how much time one could really spend continuously on working on AGI - refer to the Princeton Instititue of Advanced Studies where established geniuses (such as Einstein) were/are paid to devote fulltime efforts to thinking but actually not many earthshaking ideas have come out of it. Don't we need a lot of 'time wasted' on trivia such as a real job, leaking plumbing and family in order to have these 1 or 2 hours of creative thinking/work each day? Would it help to have consolidated 8 hour or longer blocks each day? Do people like Ben, Leitz and Peter (Voss) really have so much time to think creatively/design or is my suspicion right that a lot of their (your :) time is spent on fundraising, PR, communication, management? I think I personally spend about 30 hours/week actively focused on AGI, these days. However, the rest of my work time is spent doing activities that help bring in the $$ that pays other members of the Novamente team to work on AGI. We do have several team members working full-time on AGI RD. My total work time is probably about 65 hours a week all total, on average, though of course for much of the remainder of the week my mind is still churning about AGI and other related scientific and technology issues! I would of course like to see things shift so that I could spend, say, 50 instead of 30 hours per week on AGI directly. But as things are now, I am the business leader of Novamente LLC as well as the head AGI guru. -- 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/?member_id=231415user_secret=fabd7936
[agi] RE: Time Enough For Work
Hi John, Could you tell us more about WKG - Web Knowledge Gatherer? - Original Message - From: John G. Rose [EMAIL PROTECTED] To: agi@v2.listbox.com Sent: Monday, May 07, 2007 2:02 PM Subject: **SPAM** RE: Time Enough For Work [WAS Re: [agi] What would motivate you to put work into an AGI project?] I've found that if I can do 16 to 24 hours on, and then sleep, and then another 16 to 24 hours on as long as the body can keep up I can reach really high thresholds of productivity with ephemeral visions of deep, comprehensive insight. In the past I've done self-directed data compression RD, written video games, telephony switching software in this way on up to 2 year stints. 8 hours is just warming up. And the concept of the Long Day Society I've kicked about for years where the day I think is too short at 24 hours. Stretching the wake sleep cycle could help us live longer But the software I'm working on (WKG - Web Knowledge Gatherer) is pre-AGI and at some point will need to grow a brain :) so reading all these discussions and interactions especially among the more astute and learned individuals on this email list is very informative gives perspective on some of the RD that is going on. And any posts and references on good learning material and books are helpful. I've just started reading The Symbolic Species by Terrence W. Deacon which has sat on my shelf for years :) and is a little outdated I suppose but has some good information and provides some examples examining the human brain but ... brain and computer software s different and the brain is such a conglomerated mish-mash of evolutionary cognitive appendages! It's almost like OK need to start from scratch when building AGI like when software becomes fragile, rigid, brittle, and rots (as Agile design tries to avoid). If the brain could be decompiled, which I'm sure we are getting closer, how much of it would really be useful for AGI? Would the source code be too messy and spaghetti like? Are there any algorithms and data structures that haven't been discovered in mathematics? And modeling AGI after brain... maybe loosely. Do we model machines after human body design, some yes but others not, say an army tank is in some ways like a human body but it more accommodates humans (and disaccommodates) than is modeled after. The decompiled brain source code would have some really amazing undiscovered stuff in there. Maybe it couldn't be represented with conventional programming languages. I wonder... parts of it would be immensely sophisticated yes that's a de facto assumption :). And it could be decompiled at many levels. I suppose a functional level, macroscopic decompiler could be made now there are probably many in existence... John -Original Message- From: Richard Loosemore [mailto:[EMAIL PROTECTED] Sent: Monday, May 07, 2007 9:58 AM To: agi@v2.listbox.com Subject: Time Enough For Work [WAS Re: [agi] What would motivate you to put work into an AGI project?] Myself, I think that the number of hours working alone might only need to be a small number (3-4). But what I value most is hours brainstorming with others who are of like mind and similar level of knowledge. That is a gold-dust situation. I have been watching From The Earth To The Moon recently. Oh to be part of a 100,000-strong community working on one noble project! Richard Loosemore. Jean-Paul Van Belle wrote: Interesting question you raise there, Matt (vs :) YKY How many of us would be prepared to work FULL-TIME on AGI: (0) If a department of defense/military organisation paid you develop a secret AGI for national defense/intelligence purpose? (1) If a Microsoft, Google, Sun or IBM came along and hired you full-time to work on either (1a) Open-Source; or (1b) Proprietary AGI? (2) A more 'friendly' research group came along (e.g. University, government agency) to pay you fulltime (2a) on *their* design/architecture or (2b) on YOUR design but having to share your findings with the larger community (shared credit)? (3) If you had sufficient funds of your own? Re (3) I have often wondered how much time one could really spend continuously on working on AGI - refer to the Princeton Instititue of Advanced Studies where established geniuses (such as Einstein) were/are paid to devote fulltime efforts to thinking but actually not many earthshaking ideas have come out of it. Don't we need a lot of 'time wasted' on trivia such as a real job, leaking plumbing and family in order to have these 1 or 2 hours of creative thinking/work each day? Would it help to have consolidated 8 hour or longer blocks each day? Do people like Ben, Leitz and Peter (Voss) really have so much time to think creatively/design or is my suspicion right that a lot of their (your :) time is spent on fundraising, PR, communication, management? The grass always seems greener on the other side... Jean-Paul -
RE: [agi] RE: Time Enough For Work
WKG is an application that takes spidering a few other common technologies and ties them together. Nothing really revolutionary... this is the 3rd prototype since 2000. The first was written in Visual Basic 6, the second in Delphi in 2002, and this version is a combo of C# and C++. This version will go on the market barring any issues. I'm aiming to have a beta in about 4 months depending on how things go. WKG is one of those things that as soon as it is released it will get imitated so I can't say too much on what it is and have to push the technology as far as I can before release... also it may just be ignored or used by just a few rare individuals and organizations. Either way I'm writing it and am compelled to get it done since no-one has anything exactly like it or has done it yet. So this doesn't say much, but WKG relies on the state of current technology where we have fast computers and fast internet cheap, with large chunks of the internet available for free within millisecond reach. This open internet may not be around forever and now is the time to somehow utilize the situation... Also with all this extremely valuable information building a text only brain, not including multi-media, is next on my list. The internet is rapidly changing now with video and interpreting video is a whole 'nother world... but a text processing based brain, where is one? Is it that difficult? With all this advanced linguistics technology freely available a text brain has got to be a no brainer :) John From: Mark Waser [mailto:[EMAIL PROTECTED] Hi John, Could you tell us more about WKG - Web Knowledge Gatherer? - 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?
On 5/7/07, James Ratcliff [EMAIL PROTECTED] wrote: One goal or project I was considering (for profit) is a research tool, basically a KB that scans in teh newspapers and articles and extracts pertinent information for others to query against and use. This would help build up a large world knowledge base, and would also be salable to research companies and such. One example of that is the tragedy shooting at VT this past month, I ran some scritps against the news article and came up with a lot of hidden information in there about the Chu guys family and some other conenctions that I wasnt seeing in many of the news articles, that let me go down some other paths to find info. Another goal or application was a 3D avatar bot like Novamente is now pursuing. This could be used most easily to simulate an autonomous AGI agent that could act in a 3d rich world. I think web page classification is a good first goal for AGI. Though there may be competition from other newer search engines such as PowerSet. For the 3D avatar, it seems very difficult to commercialize (but I may be ignorant of areas like gaming or Second Life). 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/?member_id=231415user_secret=fabd7936
Re: Time Enough For Work [WAS Re: [agi] What would motivate you to put work into an AGI project?]
Ben is about the most productive and energetic AGI person I've seen =) Personally I think I'm better at theorizing than software development, and my work/sleep hours are so irregular that I can't keep account of them, except to say that I'm devoted to AGI full-time. BTW.. I think churning code efficiently is a skill that is complementary to mine, and is not meant to be derogatory. AGI probably requires many different personalities working together. 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/?member_id=231415user_secret=fabd7936