[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference rusure::math

Title:Mathematics at DEC
Moderator:RUSURE::EDP
Created:Mon Feb 03 1986
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:2083
Total number of notes:14613

1619.0. "A 'wobbly' sequence" by TRACE::GILBERT (Ownership Obligates) Tue Jun 02 1992 17:56

    A sequence has the property that the sum of any 7 consecutive elements
    is positive, while the sum of any 11 consecutive elements is negative.
    What is the longest such sequence?
T.RTitleUserPersonal
Name
DateLines
1619.1Do I get a cookie? :-)GUESS::DERAMODan D'Eramo, zfc::deramoTue Jun 02 1992 22:199
>    A sequence has the property that the sum of any 7 consecutive elements
>    is positive, while the sum of any 11 consecutive elements is negative.
>    What is the longest such sequence?
        
        I have a proof that the longest such sequence is less
        than seventy-seven elements, but it is too long to fit
        into the margin.
        
        Dan
1619.2DESIR::BUCHANANWed Jun 03 1992 11:2910
>    A sequence has the property that the sum of any 7 consecutive elements
>    is positive, while the sum of any 11 consecutive elements is negative.
>    What is the longest such sequence?
        
	Where did you come across this puzzle?   I've remet two of the other 
five in technical AI journals, where "automated reasoning techniques" were 
being applied.

Nostalgically,
Andrew.
1619.3a solutionDESIR::BUCHANANMon Jun 08 1992 11:3243
	A solution after the form-feed...



	Let's call our sequence A_1, A_2, ... A_n.
	And let's define S_0 = 0
			 S_j = sum (i=1...j) A_i

	Now, the properties of adjacent terms can be expressed as:
		S_j < S_(j+7) for all j in {0,...,n-7}
		S_k < S_(k-11) for all k in {11,...,n}	

	Suppose now that n >= 17.

	S_17 < S_6 < S_13 < S_2 < S_9 < S_16 < S_5 
	< S_12 < S_1 < S_8 < S_15 < S_4 < S_11
	< S_0 < S_7 < S_14 < S_3 < S_10 < S_17

	which is impossible.   So n =< 16.   In fact, n = 16 *is* attainable.
By setting the values of S_j to anything satisfying:

	       S_6 < S_13 < S_2 < S_9 < S_16 < S_5 
	< S_12 < S_1 < S_8 < S_15 < S_4 < S_11
	< S_0 < S_7 < S_14 < S_3 < S_10

where S_0 = 0, we can compute immediately the values of A_j, since the 
definition of S in terms A is invertible.   So for instance if S_6 = -12, 
S_13 = -11, S_2 = -10, ... , S_0 = 0, ... , S_10 = 4 then we have

	A_j = 13 if j = 3,7,10 or 14
	A_j = -5 otherwise.

Every subsequence of length 7 sums to 1, and every subsequence of length 11
sums to -1.

-------------------------------------------------------------------------------

	We can generalize the problem from 7 & 11 to p & q.   Now we find that
the maximum n is given by:
		p + q - (p,q) - 1
where (p,q) is the highest common factor of p & q.

Andrew