> > 2. I believe that AGI can’t move forward until programs have the > unlimited ability to create other programs. This hasn’t happened yet and > therefore I have no proof that this capability will be more successful than > current approaches BUT at least it isn’t one of the techniques that we know > doesn’t work. > **** > My guess is that this isn't necessary. As long as they can change their > internal maps and structures in the data -- that should be close enough to > creating from-scratch programs to solve AI/AGI problems. But, one > computer's data is another computer's program.
There's also Lisp (where every program is a nested list, and any properly formed nested list can therefore be executed as a program, one of the reasons Lisp has always been touted as an AI language), C# and Python (where a program can create a string and request it to be byte-compiled and executed as a program), and other interpreted or byte-compiled languages that work along similar lines. I don't know of any offhand, but I'm sure there are machine code-compiled languages that have a similar feature. This isn't a new feature to programming languages, it's just not a commonly used one. On Thu, Jan 10, 2013 at 2:48 PM, Dimitry Volfson <[email protected]> wrote: > A couple comments below: > > > On 1/10/2013 11:54 AM, David Clark wrote: > > This response is for Aaron and John.**** > > ** ** > > Both, however, are mathematical. You are shadow boxing.**** > > ** ** > > Again, conclusions with no supporting arguments.**** > > ** ** > > Although I don’t put much value in “beliefs” or “intuition”, I am guilty > of a few as well.**** > > ** ** > > 1. I believe that AGI can be created using software we have today, > on computers that exist today. AGI has not been created yet so I have no > proof that this is true, even though I am acting as if it is. > > Probably true. Some problems can be solved with much older computers just > because we have many more training examples now than ever before -- that > can be digested by a lot of processing and the digested info can be passed > along to lesser power computers. And, the computing power that Watson > requires can be had for just $200 per hour from Amazon's service, so anyone > can rent a supercomputer. > > **** > > 2. I believe that AGI can’t move forward until programs have the > unlimited ability to create other programs. This hasn’t happened yet and > therefore I have no proof that this capability will be more successful than > current approaches BUT at least it isn’t one of the techniques that we know > doesn’t work.**** > > > My guess is that this isn't necessary. As long as they can change their > internal maps and structures in the data -- that should be close enough to > creating from-scratch programs to solve AI/AGI problems. But, one > computer's data is another computer's program. > > -- Dimitry Volfson > *AGI* | Archives <https://www.listbox.com/member/archive/303/=now> > <https://www.listbox.com/member/archive/rss/303/23050605-2da819ff> | > Modify<https://www.listbox.com/member/?&;>Your Subscription > <http://www.listbox.com> > > > ____________________________________________________________ > *Woman is 53 But Looks 25* > Mom reveals 1 simple wrinkle trick that has angered doctors... > <http://thirdpartyoffers.juno.com/TGL3142/50ef2935323102934059dst03vuc> > ConsumerLifestyleMag.com<http://thirdpartyoffers.juno.com/TGL3142/50ef2935323102934059dst03vuc> ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
