Jim, It is hard to understand what you are suggesting. But I get a hunch you may be saying something like the idea I have been promoting for a long time. This is that for cognition all or most compressions will at best be partial (lossy?)
I suppose you could keep a list of outlier data for each partial compression, and adjust processing based on it. Is that what you are suggesting? My guess is that we won't gain that much by compressing at the end of the day anyway. Processing distributed data in parallel will probably be as efficient as processing compressed data in serial. So my guess is it will be better to just work with uncompressed data, in parallel. But I like the insight that all compressions will be partial. If that is what you are suggesting. If that is what you are suggesting I agree with you. In short, I think the best way to address the partial/subjective compression problem I believe you to be addressing, will be ad-hoc compressions relevant to specific decisions. -Rob On Wed, Jun 7, 2017 at 3:16 AM, Jim Bromer <[email protected]> wrote: > *A program could create new rules at a higher abstraction level that could > operate on the rules that exist at a lower level. (This characterization is > relative but I think it is useful.) This might allow for greater > compressive individuation.* > > *So creating individuated sub programs based on the data that was being > compressed (or transformed from one compressed form to another) might be > useful in devising ways to effectively work with the data without fully > decompressing it.* > > *Logic is (typically) a compressed representation and you can often work > with in compressed form. But, there are many common situations where that > data has to be excessively decompressed in order to work with it. My > conjectured-goal is for a program to be able to create new rules during an > analytical run so that it could individuate the data (to compress it) and > still be able to operate on the different individuations without fully > decompressing them.**Even if I can’t get it to work, I should be able to > show that it can work on some special cases.* > > > Jim Bromer > > On Tue, Jun 6, 2017 at 10:02 AM, Jim Bromer <[email protected]> wrote: > >> I meant: >> This would allow for greater individuation which is necessary in order to >> work with the data in compressed form because the data would be too >> extensive for the program to handle it in decompressed form. >> >> Jim Bromer >> >> On Tue, Jun 6, 2017 at 10:00 AM, Jim Bromer <[email protected]> wrote: >> >>> Most programs operate on a higher level of rules which operate on the >>> data that is encountered. This higher level program is a kind of general >>> abstraction. A program can create new rules at higher abstraction that >>> could operate on the rules at a lower level. (This characterization is >>> relative but I think it is useful. Programming languages do this but the >>> essential part of the program produced is created by a person) This would >>> allow for greater individuation which is necessary in order to work with >>> the data in decompressed form because the data would be too extensive for >>> the program to handle it in decompressed form. This individuation would >>> mean that the data could not be decompressed without the higher level >>> abstractions (or sub programs) that each individual run would produce. >>> These higher level sub-programs would be part of the data the program would >>> create but I am trying to emphasize the idea that the program would create >>> sub-programs based on the data that it was encountering and compressing. >>> (My idea of working with SAT is that it would transform statements in >>> traditional Boolean Logical form into another form which would also be >>> compressed.) >>> So creating individuated sub programs based on the data that was being >>> compressed (or transformed from one compressed form to another) might be >>> useful in devising ways to compress the data. >>> This kind of program would use generalizations as components both to >>> directly represent the data in compressed form and to create sub-programs >>> that could operate on the compressed data. >>> >>> Jim Bromer >>> >>> On Mon, Jun 5, 2017 at 10:04 PM, Jim Bromer <[email protected]> wrote: >>> >>>> I realized that the encoding example was not a very good one because a >>>> compression method has to include a way to decompress the data to produce >>>> the reference object. Suppose you used generalizations which either >>>> referred to elaborate data objects or which could be elaborated more fully >>>> based on context then that might be used in a system that could operate (in >>>> some way) on the compressed data without decompressing it. If the >>>> compression data/operations method was able to create specialized >>>> compression systems based on the data encountered, then different threads >>>> or computers would develop individuated data systems. Different threads or >>>> computers might then be able to talk to each other in order to provide >>>> information about what they encountered without fully decompressing the >>>> data. They would have to have conversations to explain themselves. Suppose >>>> that they used some common basis of reference to start with, but then >>>> developed individuated systems. And further suppose that the compression >>>> method was like a language. While their conversation would represent a >>>> partial elaboration of the data that they had acquired, they would not have >>>> to fully decompress the information they had in order to explain the >>>> information they acquired to the other computer in spite of the fact that >>>> their records of information were compressed in different ways. >>>> >>>> This idea of being able to operate on compressed representations >>>> without having to fully decompress them might explain why my visual >>>> recognition memory is good enough even though my visual recollection >>>> is very unsophisticated (in spite of my knowing a lot about painting.) >>>> >>>> Jim Bromer >>>> >>>> On Mon, Jun 5, 2017 at 6:24 PM, Mike Archbold <[email protected]> >>>> wrote: >>>> >>>>> It seems like everything is "compressed" some way usually in computer >>>>> processing. Going back to my banking days we'd have a debit to >>>>> general ledger, but what this represents is somebody walking around >>>>> and filling out a ticket, perhaps talking to someone else, but all the >>>>> software sees is the end result which could be viewed as a kind of >>>>> compression of the entire transaction. "I went to Europe" is a >>>>> compression of the trip down to a single proposition. But I think >>>>> what you are asking is: can programming be done in generalizations, >>>>> and if so, how can that be formalized? >>>>> >>>>> On 6/5/17, Jim Bromer <[email protected]> wrote: >>>>> > An encoding is almost always a compression method. The data encoding >>>>> is >>>>> > referring to some kind of object or event which can be described more >>>>> > fully. So anytime we devise or use an encoding and a system of >>>>> operations >>>>> > that can act on those encoded references we are effectively >>>>> developing a >>>>> > compression system that can act on some kind of compressed data >>>>> without >>>>> > fully (or excessively) decompressing it. >>>>> > >>>>> > So the basis for this kind of thing is well established. >>>>> > >>>>> > Jim Bromer >>>>> > >>>>> > On Mon, Jun 5, 2017 at 6:31 AM, Jim Bromer <[email protected]> >>>>> wrote: >>>>> > >>>>> >> I realized that traditional logic is a system of compression which >>>>> allows >>>>> >> for some computations that can be run without fully decompressing >>>>> the >>>>> >> data. >>>>> >> However, at certain steps at some (relatively low) levels of >>>>> complexity >>>>> >> the >>>>> >> data has to be decompressed (to a great degree). So this example >>>>> proves >>>>> >> the >>>>> >> system is feasible and it is not completely based on binary >>>>> addition or >>>>> >> multiplication methods (which are also examples of compression >>>>> systems >>>>> >> which can operate on compressed data without decompression.) I did >>>>> not >>>>> >> want >>>>> >> to use binary arithmetic as an example because computers were >>>>> designed >>>>> >> around those principles. >>>>> >> >>>>> >> So since an example is easy to find, this proves that the >>>>> methodology can >>>>> >> be studied as a separate branch of computer science. The question >>>>> then is >>>>> >> whether other, more powerful systems, can be developed. >>>>> >> >>>>> >> Jim Bromer >>>>> >> >>>>> >> On Sun, Jun 4, 2017 at 4:42 PM, Jim Bromer <[email protected]> >>>>> wrote: >>>>> >> >>>>> >>> The Halting Problem shows that the results of programs >>>>> (programmable >>>>> >>> logic) cannot be completely computed - for every possible program - >>>>> >>> without >>>>> >>> running the program. (The Gödel Incompleteness Theorem shows that >>>>> there >>>>> >>> are -some- comprehendible logical problems that could not even be >>>>> >>> theoretically resolved programmatically.) >>>>> >>> >>>>> >>> But there are some programs that can be computationally processed >>>>> so >>>>> >>> that >>>>> >>> a system of results can be produced more quickly than actually >>>>> running >>>>> >>> the >>>>> >>> program. >>>>> >>> >>>>> >>> The significance of this came to me after I started criticizing the >>>>> >>> proposition that an advanced representational compression could >>>>> >>> be sufficient to produce AGI at this time. The problem is that >>>>> >>> representational compressions have to go through stages of >>>>> decompression >>>>> >>> and recompression in order to do any computation on the data, and >>>>> given >>>>> >>> the >>>>> >>> degree of compression that would be needed for AGI that would make >>>>> the >>>>> >>> system way too slow. >>>>> >>> >>>>> >>> While logical computation is a simple process using binary >>>>> >>> representations of simple logical states for each literal (logical >>>>> >>> variable), the problem is that logical Satisfiability statements >>>>> are >>>>> >>> (most >>>>> >>> familiarly) compressions of multiple logical states. A logical >>>>> statement >>>>> >>> is >>>>> >>> (typically) a compression of a system of individual 'solutions'. So >>>>> >>> what would be needed would be a computational method that can act >>>>> >>> efficiently on a wide variety of logical solutions. In other words >>>>> we >>>>> >>> need >>>>> >>> a compression method which can do computations on the compressed >>>>> >>> representations without (excessively) decompressing them for each >>>>> >>> computation. I started thinking about this project from the view >>>>> that my >>>>> >>> goal is not to make p=np but to try to develop a methodology that >>>>> might >>>>> >>> one >>>>> >>> day be more efficient than methods that we currently use. >>>>> >>> >>>>> >>> I started thinking that it might be possible to create compression >>>>> >>> representations that acted from different levels of abstraction. >>>>> These >>>>> >>> levels of abstraction might be thought of as programs. I started >>>>> >>> thinking >>>>> >>> about Turing's Halting Problem and I realized that while you >>>>> cannot find >>>>> >>> a >>>>> >>> shortcut to the completion state of every possible computer >>>>> program, >>>>> >>> you can for some kinds of programs. The results of a system that I >>>>> >>> am imagining could be decompressed (at the end of the analytical >>>>> >>> computational stages) to produce solutions using the >>>>> >>> levels-of-abstraction >>>>> >>> sub-program. but the system could run through the computational >>>>> steps >>>>> >>> without actually running that sub-program. The compressed >>>>> >>> representations >>>>> >>> would not have to be excessively decompressed in order to run a >>>>> >>> computation >>>>> >>> on them. >>>>> >>> >>>>> >>> I am all but certain that an example is feasible (although I do >>>>> not have >>>>> >>> one right now). >>>>> >>> >>>>> >>> And incidentally, an advancement that shows that a compression >>>>> system >>>>> >>> might be accompanied by an effective computational method that can >>>>> act >>>>> >>> on >>>>> >>> the compressed representations without fully decompressing them >>>>> might be >>>>> >>> interesting to some people. The system does not have to beat >>>>> Turing at >>>>> >>> run-around-the-house chess or to prove Gödel wrong from under a >>>>> table at >>>>> >>> a >>>>> >>> week-long Oktoberfest or be dependent on winning a million >>>>> dollars, it >>>>> >>> only >>>>> >>> has to be interesting, incrementally improvable and salient to the >>>>> study >>>>> >>> of >>>>> >>> logic. >>>>> >>> Jim Bromer >>>>> >>> >>>>> >> >>>>> >> >>>>> > >>>>> > >>>>> > >>>>> > ------------------------------------------- >>>>> > AGI >>>>> > Archives: https://www.listbox.com/member/archive/303/=now >>>>> > RSS Feed: https://www.listbox.com/member >>>>> /archive/rss/303/11943661-d9279dae >>>>> > Modify Your Subscription: >>>>> > https://www.listbox.com/member/?&; >>>>> > Powered by Listbox: http://www.listbox.com >>>>> > >>>>> >>>>> >>>>> ------------------------------------------- >>>>> AGI >>>>> Archives: https://www.listbox.com/member/archive/303/=now >>>>> RSS Feed: https://www.listbox.com/member/archive/rss/303/24379807-6537 >>>>> 94b5 >>>>> Modify Your Subscription: https://www.listbox.com/member >>>>> /?& <https://www.listbox.com/member/?&;> >>>>> Powered by Listbox: http://www.listbox.com >>>>> >>>> >>>> >>> >> > *AGI* | Archives <https://www.listbox.com/member/archive/303/=now> > <https://www.listbox.com/member/archive/rss/303/29034623-d94dbe32> | > Modify > <https://www.listbox.com/member/?&;> > Your Subscription <http://www.listbox.com> > ------------------------------------------- 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
