Arbitrary array of numbers, proof

AI Thread Summary
The discussion revolves around proving that in a 5x5 array of numbers from 1 to 25, the least of the greatest numbers in each row (s) is always greater than or equal to the greatest of the least numbers in each column (t). Participants suggest starting the proof by assuming the opposite, that s is less than t, and exploring the implications. This leads to a contradiction, as it shows that t cannot be the least in its column if it is greater than s. Ultimately, the conclusion is reached that s must be greater than or equal to t, confirming the initial claim. The proof effectively utilizes the properties of the array's arrangement to establish the relationship between s and t.
Daltohn
Messages
30
Reaction score
0

Homework Statement


The numbers 1 to 25 are arranged in a square array of five rows and five columns in an arbitrary way. The greatest number in each row is determined, and then the least number of these five is taken; call that number s. Next, the least number in each column is determined, and then the greatest number of these five is taken; this number is called t.
Prove that s is equal to or greater than t for all possible arrangements.

Homework Equations

The Attempt at a Solution


I see intuitively that this is the case but have no idea where to start with a proof.
 
Physics news on Phys.org
How about reversing the problem ? assume t > s and show that it leads to a contradiction ...
 
Suppose it is false, i.e. s<t. What can you say about other numbers in the same row as s?
 
  • Like
Likes BvU
haruspex said:
Suppose it is false, i.e. s<t. What can you say about other numbers in the same row as s?
I don't know except that they're all less than s. What more can I deduce assuming s<t?
 
s is the least of the biggers and t is the greatest of the smallers, right?
Say, ## s = \min( s_1, s_2, s_3, s_4, s_5) ## where ##s_n ## denotes the largest member in row n.
Similarly, ## t= \max( t_1, t_2, t_3, t_4, t_5) ## where ##t_m ## denotes the smallest member in column m.
You know something about row n where ##s = s_n##.
Daltohn said:
I don't know except that they're all less than s.
Use this knowledge to contradict the assumption that s<t.
Each of the t_m terms are the minimums from their columns.
 
Okay, I'm an idiot...

Suppose s<t. Then t is not in the same row as s since none of its members is greater than s. But then t can never be the least in its column since every column has a member that is less than or equal to s. However, t is by definition the least in its column. Contradiction. Hence s is greater than or equal to t.

Something like that? Maybe the contradiction is that t both is and isn't a member of the same row as s.
 
You've got it. The best you can hope for is t=s.
 
Yep, thank you all :)
 
Back
Top