Google Answers Logo
View Question
 
Q: Upper bound of Kolmogorov complexity in concatenation ( No Answer,   5 Comments )
Question  
Subject: Upper bound of Kolmogorov complexity in concatenation
Category: Computers > Algorithms
Asked by: nr9-ga
List Price: $100.00
Posted: 03 Aug 2004 12:19 PDT
Expires: 04 Aug 2004 04:56 PDT
Question ID: 383049
How do you show that for any c, some strings x and y exist, where
K(xy) > K(x) + K(y) + c?

In other words, show that the upper bound of K(xy) cannot be improved
to K(x) + K(y) + c
Answer  
There is no answer at this time.

Comments  
Subject: Re: Upper bound of Kolmogorov complexity in concatenation
From: joshfraz-ga on 03 Aug 2004 15:46 PDT
 
Hey there,

You would expect that for all strings x and y:
K(xy) <= K(x) + K(y) + C, for some C.  However, this is not true.  

The problem lies in the seperation of d(x) and d(y),  Instead we have
a constanct C such that: K(xy) <= K(x) + K(y) + 2log(K(x)) + C, for
all strings x and y.

Proof:

Let m be the logarithm of |d(x)|, then the string ?1m0 <|d(x)|> d(x)?
gives a self-delimiting description of x.  (We need 2m bits to
indicate the length of d(x).)  Hence the input ?1m0 <|d(x)|> d(x)
d(y)? gives an unambiguous description of xy.

I hope that helps!  

Josh Fraser
Subject: Re: Upper bound of Kolmogorov complexity in concatenation
From: nr9-ga on 03 Aug 2004 16:03 PDT
 
sure, but that doesnt prove we can't improve on it, does it?

Just because we can't construct an representation of xy that yields
K(x) + K(y) + c doesn't mean it doesn't exist.
Subject: Re: Upper bound of Kolmogorov complexity in concatenation
From: maniac-ga on 03 Aug 2004 16:52 PDT
 
Hello Joshfraz,

Hmm. I never even heard about Kolmogorov complexity before this
question so I won't hesitate to make an answer [unless you ask for
one]. However this certainly sounds like a problem where a proof by
induction would fit the bill.

[1] Based on what you know, can you construct a string x and y such that
  K(xy) > K(x) + K(y) + 1

[2] If yes, and assuming you have a string x and y such that
  K(xy) > K(x) + K(y) + N
can you construct a string xa and yb such that
  K(xayb) > K(xa) + K(yb) + N+1

Based on the comment made by joshfraz, it would seem that extending
the strings x to xa and y to yb by some amount (perhaps the length of
a and b should be based on a function of N) to make the second
relationship true.

Then using induction, you can prove the original statement. If this
works for you, I'd suggest adjusting the list price to a more
appropriate value so I can post a proper answer. If not, perhaps one
of our other researchers can help you.

  --Maniac
Subject: Re: Upper bound of Kolmogorov complexity in concatenation
From: nr9-ga on 03 Aug 2004 17:27 PDT
 
I don't think kolmogorov complexity can be computed like that

http://www3.oup.co.uk/computer_journal/hdb/Volume_42/Issue_04/
Subject: Re: Upper bound of Kolmogorov complexity in concatenation
From: nr9-ga on 03 Aug 2004 17:44 PDT
 
the high price is because I think this is a difficult problem to solve.

I don't need a well researched, eloquent answer, just a concise proof
that I can understand.

Important Disclaimer: Answers and comments provided on Google Answers are general information, and are not intended to substitute for informed professional medical, psychiatric, psychological, tax, legal, investment, accounting, or other professional advice. Google does not endorse, and expressly disclaims liability for any product, manufacturer, distributor, service or service provider mentioned or any opinion expressed in answers or comments. Please read carefully the Google Answers Terms of Service.

If you feel that you have found inappropriate content, please let us know by emailing us at answers-support@google.com with the question ID listed above. Thank you.
Search Google Answers for
Google Answers  


Google Home - Answers FAQ - Terms of Service - Privacy Policy