A problem about differentiability

  • Thread starter Thread starter rasi
  • Start date Start date
  • Tags Tags
    Differentiability
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
2 replies · 1K views
rasi
Messages
18
Reaction score
0
i tried to solve this problem. i can do it a little. but i can't progress. as far as I'm concerned, it requires outstanding performance. thanks for now...
PROBLEM
5625.jpg

MY SOLUTION...
5625_ms_s.jpg

 
Physics news on Phys.org
This is pretty tricky (unless I missed something clever). Think about this geometrically: we know that f(x) is oscillating between -1 and 1 in some fashion. The statement
[tex]f'(x)^2+f''(x)^2 \leq 1[/tex]
says that the derivative of f and the second derivative cannot both be big at the same time. So for example if f'(x) is big when f(x) is near 1, it won't be able to decrease fast enough (because f''(x) is small) to prevent f(x) from crossing the value of 1, which gives a contradiction.

The question then is how to put this into something more rigorous. Suppose we're at a point a for which f(a)2+f'(a)2>1. The picture you should have in your head (not something you should assume but something to give you an idea of wha's going on) is f(a) and f'(a) both positive. Then f is going to eventually reach the value of 1 unless f'(x) decreases and is eventually 0. So the question that you need to answer: how fast can f'(x) decrease to reach 0, given that f'(x)2+f''(x)2<1?