Hello Skeletori, Welcome to the list. I enjoy your comments and rationalization regarding personal identity and of why we should consider I to be the universe / multiverse / or the everything. I have some comments regarding the technological singularity below.
On Sat, Apr 3, 2010 at 5:23 PM, Skeletori <sami.per...@gmail.com> wrote: > Hello! > > I have some tentative arguments on TS and wanted to put them somewhere > where knowledgeable people could comment. This seemed like a good > place. I also believe in an ultimate ensemble but that's a different > story. > > Let's start with intelligence explosion. This part is essentially the > same as Hawkins' argument against it (it can be found on the Wikipedia > page on TS). > > When we're talking about self-improving intelligence, making improved > copies of oneself, we're talking about a very, very complex > optimization problem. So complex that our only tool is heuristic > search, making guesses and trying to create better rules for taking > stabs in the dark. > > The recursive optimization process improves by making better > heuristics. However, an instinctual misassumption behind IE is that > intelligence is somehow a simple concept and could be recursively > leveraged not only descriptively but also algorithmically. If the > things we want a machine to do have no simple description then it's > unlikely they can be captured by simple heuristics. And if heuristics > can't be simple then the metasearch space is vast. I think some people > don't fully appreciate the huge complexity of self-improving search. > > The notion that an intelligent machine could accelerate its > optimization exponentially is just as implausible as the notion that a > genetic algorithm equipped with open-ended metaevolution rules would > be able to do so. It just doesn't happen in practice, and we haven't > even attempted to solve any problems that are anywhere near the > magnitude of this one. > > So I think that the flaw in IE reasoning is that there should, at some > higher level of intelligence, emerge a magic process that is able to > achieve miraculous things. > > If you accept that, it precludes the possibility of TS happening > (solely) through an IE. What then about Kurzweil's law of accelerating > returns? Well, technological innovation is similarly a complex > optimization problem, just in a different setting. We can regard the > scientific community as the optimizing algorithm here and come to the > same conclusions as with IE. That is, unless humans possess some kind > of higher intelligence that can defeat heuristic search. I don't think > there's any reason to believe that. > > Complex optimization problems exhibit the law of diminished returns > and the law of fits and starts, where the optimization process gets > stuck in a plateau for a long time, then breaks out of it and makes > quick progress for a while. But I've never seen anything exhibiting a > law of accelerating returns. This would imply that, e.g., Moore's law > is just "an accident", a random product of exceedingly complex > interactions. It would take more than some plots of a few data points > to convince me to believe in a law of accelerating returns. If not the plots what would it take to convince you? I think one should accept the law of accelerating returns until someone can describe what accident caused the plot. Kurzweil's page describes a model and assumptions which re-create the real-world data plot: http://www.kurzweilai.net/articles/art0134.html?printable=1 It is a rather long page, Ctrl+F for "The Model considers the following variables:" to find where he describes the reasoning behind the law of accelerated returns. > It also > depends on how one defines exponential growth, as one can always take > X as exp(X) - I suppose we want the exponential growth of some > variable that is needed for TS and whose linear growth corresponds to > linear increase in "technological ability" (that's very vague, can > anybody help here?). > > In conclusion, I haven't yet found a credible lawlike explanation of > anything that could cause a "runaway" TS where things become very > unpredictable. > > All comments are welcome. > I think intelligence optimization is composed of several different, but interrelated components, and that it makes sense to clearly define these components of intelligence rather than talk about intelligence as a single entity. I think intelligence embodies. 1. knowledge - information that is useful for something 2. memory - the capacity to store, index and organize information 3. processing rate - the rate at which information can be processed The faster the processing rate, the faster knowledge can be applied and the faster new knowledge may be acquired. There are several methods in which new knowledge can be be generated: Searching for patterns and relations within the existing store of knowledge (data mining). Proposing and investigating currently unknown areas (research). Applying creativity to find more useful forms of knowledge (genetic programming / genetic algorithms). All three of these methods accelerate given a faster processing rate. Consider for example, our knowledge of protein folding. Our knowledge of it is almost entirely dependent on our ability to process information. I don't think anyone would argue that the amount knowledge possessed by our civilization is not increasing. If the physical laws of this universe are deterministic then there is some algorithm describing the process for an ever increasing growth in knowledge. Some of this knowledge may be applied toward creating improved versions of memory or processing hardware. Thus creating a feed-back loop where increased knowledge leads to better processing, and better processing leads to an accelerated application and generation of knowledge. I think it is easy for one's intuition to get stuck, when considering the possibility of something like a (programming language) compiler which is so good at optimization that when run against its own code, it will create an even more optimized form of itself, which could in turn make an even better version of itself, ad infinitum. I think the difficulty in imagining this is that it assumes only one piece of the puzzle, I think in this case, knowledge of building a better compiler. Knowledge of building a better compiler alone can't generate any new information about building the next best compiler. Lets take a different example, a genetic algorithm which optimizes computer chip design, forever searching for more efficient and faster hardware designs. After running for some number of generations, the most fit design is taken, assembled, and the the software is copied to run on that new hardware. Would the rate of evolution on this new, faster chip not exceed the previous rate? A more human example would be that some scientists discovered some genes that affected intelligence, and developed a drug to moderate those genes in a way that gave the entire population an intelligence on par with Newton or Leonardo. Would the amount of time until the next breakthrough discovery regarding human intelligence take more or less time, now that the entire populace consists of super geniuses? To active participants in the process, it would never seem that intelligence ran away, however to outsiders who shun technology, or refuse to augment themselves, I think it would appear to run away. Consider at some point, the technology becomes available to upload one's mind into a computer, half the population accepts this and does so, while the other half reject it. On this new substrate, human minds could run at one million times the rate of biological brains, and in one year's time, the uploaded humans would have experienced a million years worth of experience, invention, progress, etc. It would be hard to imagine what the uploaded humans would even have in common or be able to talk about after even a single day's time (2,700 years to those who uploaded). In this sense, intelligence has run away, from the perspective of the biological humans. A deeper question is what is the upper limit to intelligence? I haven't yet mentioned the role of memory in this process. I think intelligence is bound by the complexity of the environment. From within the computer, new, more complex environments can be created. (Just think how much more complex our present day environment is than 200 years ago), however the ultimate limit of the complexity of the environment that can be rendered depends on the amount of memory available to represent that environment. Evolution to this point has leveraged the complexity of the physical universe and the presence of other evolved organisms to create complex fitness tests, but evolution would hit a wall if it reached a point where DNA molecules couldn't get any longer. Jason -- You received this message because you are subscribed to the Google Groups "Everything List" group. To post to this group, send email to everything-l...@googlegroups.com. To unsubscribe from this group, send email to everything-list+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/everything-list?hl=en.
