Slashdot Mirror
Text Compressor 1% Away From AI Threshold
Baldrson writes "Alexander Ratushnyak compressed the first 100,000,000 bytes of Wikipedia to a record-small 16,481,655 bytes (including decompression program), thereby not only winning the second payout of The Hutter Prize for Compression of Human Knowledge, but also bringing text compression within 1% of the threshold for artificial intelligence. Achieving 1.319 bits per character, this makes the next winner of the Hutter Prize likely to reach the threshold of human performance (between 0.6 and 1.3 bits per character) estimated by the founder of information theory, Claude Shannon and confirmed by Cover and King in 1978 using text prediction gambling. When the Hutter Prize started, less than a year ago, the best performance was 1.466 bits per character. Alexander Ratushnyak's open-sourced GPL program is called paq8hp12 [rar file]."
paq8hp12. when decompressed, it also serves as the source code for the program.
Or more usefully, compress Wikipedia onto a single SD card in my mobile phone (Palm Treo) - with SDHC format cards, it can do 8 GB today.
Compression format would need to make it possible to randomly access pages, of course, and an efficient search index would be needed as well, so it's not quite that simple.
Shouldn't AI be using lossy compression? Certainly my real intelligence uses um, where was I?
ccalam - acoustic versions of new songs.
The (unproven) idea is that if you want to do the best at guessing what comes next (similar to compression), you have to have a great understanding of how the language and human minds work, including spelling, grammar, associated topics (for example, if you're talking about the weather, "sunny" and "rainy" are more likely to come than "airplane"), and so on.
If you feed in the previous words in a conversation, the perfect compressor/predictor would know what words will come next. Such a machine could easily pass the Turing test by printing out the logical reply to what had just been stated. The idea is that the closer to the perfect compressor you have, the closer to artificial intelligence you are.
Ewige Blumenkraft.
I wonder if lossy text compression where prepositions are entirely thrown out would be effective? Based on context, your brain actually ignores a lot of words you read and fills in the blanks so-to-speak. Perhaps you can use simple grammar rules to predict which prepositions go where based on that same context?
Given that it takes something like ~17 hours (based on my rough calculations using the figures on WP) to compress 100MB of data using this algorithm on a reasonably fast computer ... I don't think you'd really want to use it for browsing from CD. No decompression figure is given but I don't see any reason why it would be asymmetric. (Although if there's some reason why it would be dramatically asymmetric, it'd be great if someone would fill me in.)
Mobile use is right out too, at least with current-generation equipment.
Looking at the numbers this looks like it's about on target for the usual resources/space tradeoff. It's a bit smaller than other algorithms, but much, much more resource intensive. It's almost as if there's an asymptotic curve as you approach the absolute-minimum theoretical compression ratio, where resources just climb ridiculously.
Maybe the next big challenge should be for someone to achieve compression in a very resource-efficient way; a prize for coming in with a new compressor/decompressor that's significantly beneath the current resource/compression curve...
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
The first poster on this topic had a good explanation - it seems like an AI problem, but not why.
Compression is about recognizing patterns. Once you have a pattern, you can substitute that pattern with a smaller pattern and a lookup table. Pattern recognition is a primary branch of AI, and is something that actual intelligences are currently much better at.
We can generally show this is true by applying the "grad student algorithm" to compression - i.e., lock a grad student in a room for a week and tell him he can't come out until he gets optimum compression on some data (with breaks for pizza and bathroom), and present the resulting compressed data at the end.
So far this beats out compression produced by a compression program because people are exceedingly clever at finding patterns.
Of course, while this is somewhat interesting in text, it's a lot more interesting in images, and more interesting still in video. You can do a lot better with those by actually having some concept of objects - with a model of the world, essentially, than you can without. With text you can cheat - exploiting patterns that come up because of the nature of the language rather than because of the semantics of the situation. In other words, your text compressor can be quite "stupid" in the way it finds patterns and still get a result rivaling a human.
Mod me down and I will become more powerful than you can possibly imagine!
1) Create a compression algorithm called the aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaaa algorithm
2) Add a long and self referencing article on wikipedia about said algorithm.
3) Use algorithm to compress first x% of wikipedia (including your own article)
4) WIN HUTTER PRIZE.
Training monkeys for world domination since 1439
This isn't true of all compression techniques, but it's true for many of them, especially advanced techniques, i.e. to compress a short video into MPEG4 can take hours, but most computers don't have a lot of trouble decompressing them in real time.
a txt spk vrsion of wiki shd fit 8gb i fink
its only becoz ppl r sch grmmr noobs tat tey nid 2 wste $
dud shd filta 2 txtspk b4 he cmpres
There, fixed that for ya.
I've been following the Hutter Prize with interest, having been into compression ever since reverse engineering Powerpacker on my Amiga 500 back in the good old days to understand how it worked (ah, happy memories).
... especially this enwik8 file which is more of a flat file dataset with a lot of unrelated terminology.
... how the hell can we possibly predict that the next words will be "Slashdot, Cigarette, Coffee". (Three subjects very close to my heart ... also my lungs, arteries, liver etc).
... the 100MB enwik8 file. A different file will need a different dictionary.
... example all the timestamps are in a very verbose character style like "2007-07-10 00:00:00" ... if we can recognize that, we could find an alternative encoding, changing 19 byte string into 32 bit long (maybe even less if we understand the epoch date he is using) ... again, "wetware" has to identify and decide this encoding right now.
... AND do all this in less than 9 hours I believe it takes for the latest compressor.
Now what just about all the compressors do, whether they are based on Neural Nets, Markov Models, Predictive Partial Matching or whatever, is to use patterns in the already seen text to predict the most likely following bit (0/1).
Now depending on the text itself, prediction based on previously seen text isn't enough
Try to predict the next word, byte or bit, when your previous text has been "Frog, Toilet, Woodwork"
Therefore some of these compressors are supplemented by a dictionary containing "useful" English words arranged so that the ones used most frequently get assigned a lower "size" of encoded string in the text pre-processor before the actual compression kicks in.
It seems that all the advances have been made on finding the optimum arrangement for this dictionary based on the text they have to process
Note also, as the enwik8 file is not truly a passage of text, more a collection of data in XML wrapper, there is also a lot to be gained simply be understanding the structure of the file itself, and finding an alternative representation for the XML components
Now for me, REAL AI would come when the compressor can actively SCAN the file to be compressed himself, recognize the file structure (be it XML, plaintext or whatever), and optimize it into a more compressible format, decide the optimum arrangment for the dictionary, decide the optimum compression technique, context orders to be used etc etc
This high bits/character rate comes at a heavy price in speed and memory, especially when good old WinZIP can get a pretty good result in a couple of minutes.
At the moment there is just too much "wetware" involvement to say this is truly AI, regardless of the bits/character rate they are achieving.
aTxtSpkVrsionOfWikiShdFit8gbIFink
itsOnlyBecozPplRSchGrmmrNoobsTatTeyNid2Wste$
dudShdFilta2TxtspkB4HeCmpres
Fixed even more.
Trolls are like broken clocks. They show the truth two times a day. The rest of the day they talk nonsense.
Actually, I can give you 100% compression already. It's just a bit lossy.
No, the most time-consuming part of most video encoders (including h.263 and h.264) is finding how the blocks have moved - searching for good matches between one frame and another. For best results, h.264 allows for the matches to not only come from the last frame, but up to the last 16! That allows for h.264 to handle flickering content much better, or situations where something is quickly covered and uncovered again e.g a person or car moving across frame, briefly covering parts of the background. Previous codecs did not handle those situations well and had to waste bandwidth redrawing blocks that were on screen just a moment prior.
The point does remain, most "compression" involves some sort of searching which is not performed when decompressing.
See? American English is actually just essentially lossless compression...
Perl: The only language that looks the same before and after RSA encryption.
Slashdot: where don knuth is an idiot because he cant grasp the awesome power of php
Now you figure out which one it was and how to decompress it.
Well, with only 256 choices, it didn't take long to check all possible decodings for one that makes sense. Ended up working for "}".
Oddly, though, the algorithm not only restored, but improved the original! I get:
"The King's English version of Wikipedia should fit in eight gigabits, I do believe. Only humanity's sphexish adherence to grammatical rules limits the attainable compression ratio; the good gentleman might wish to consider filtering to a more base patois prior to applying his algorithm".
Amazing... This discovery could single-handedly render the next generation (nearly) intelligible!