On Tue, Sep 4, 2012 at 4:06 PM, Abram Demski <[email protected]> wrote:
> Jim, > > What about the connection between prediction and compression? This is > tightly argued. The basic argument for compression as an AGI test is: > > compression<-prediction<-AGI > That is, AGI should be able to do prediction well, and predicting well > allows compression. A stronger version goes like this: > > compression<->prediction<->AGI > > That is, compressing well also allows prediction, and prediction is > central enough to AGI that we can reasonably say, predicting well would > allow AGI. > The various links and directions in this argument have different strengths > and weaknesses. My question to you is: which links are you questioning here? > > Abram > > The idea that an AGI program would act like a compressor is interesting and probably true. However, the idea that a simple decompression of compressed data explains everything about intelligence is neither interesting nor true. In asking you to believe something that is new, you are often asked to take small steps to lead you to accept a very different way of looking at things. The sense that a calculator algorithm is a thought algorithm is an important step in reconsidering the demand that there has to be a hard-edged distinction between thought and algorithms, but reaching the conclusion that this special way of thinking about things solves all the problems of creating AGI is not acceptable. And very few people in this group do accept conclusions like this. When we talk about intelligent prediction we are talking about something that is different from simple algorithmic decompression of compressed data. And there is no substantial evidence that a Levin Search form of Algorithmic Information Theory would be a solid base for AGI either. I think that if you had a way to make it tractable then there would be better ways to look at the problem. AGI might be a special class of compressors-decompressors, but, in general, an ordinary compression algorithm is not AGI. Jim ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-c97d2393 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-2484a968 Powered by Listbox: http://www.listbox.com
