I How to start with proving ##a^a + b^b > a^b + b^a##?

  • I
  • Thread starter Thread starter MevsEinstein
  • Start date Start date
  • Tags Tags
    Inequalites
MevsEinstein
Messages
124
Reaction score
36
TL;DR Summary
what the title says
When I learned about Catalan's conjecture, I morphed the expression ##3^2-2^3## to ##a^b+b^a##. Then, out of curiosity, I compared ##a^b+b^a## to ##a^a+b^b##. I have tried many numbers (greater that or equal to 1) for both ##a## and ##b##, noticing that ##a^a+b^b## is bigger than ##a^b+b^a##. I don't know how I am going to prove this since it contains unknown powers and a greater sign. What path should I consider taking to prove this inequality?
 
Mathematics news on Phys.org
I'm pretty skeptical that you can prove Catalan's conjecture with this, but let's assume ##a>b\geq 1##, and let's let ##x > y \geq 0##. We will prove##a^x + b^y \geq a^y+b^x##. Your use case is ##x=a## and ##y=b##.

##a^x+b^y=a^ya^{x-y}+b^y## and ##a^y+b^z= a^y+b^y b^{x-y}##. Both of these expressions are ##a^y+b^y## where we multiply one of the summands by some quantity. Since the summands are at least one and the multipliers are at least one, it is hopefully intuitive that if we multiply the larger summand by a larger number, the thing we get is larger overall. ##a^y>b^y## and ##a^{x-y}>b^{x-y}##, so that completes the intuitive proof.If you want to throw more algebra at it, we need to compute exactly how much larger than ##a^y+b^y## each side is
##a^ya^{x-y}+b^y = a^y(a^{x-y}-1+1)+b^y##
##=a^y(a^{x-y}-1)+a^y +b^y##

Similarly ##a^y+b^x= a^y+b^y+ b^y(b^{x-y}-1)##.

So all we have to do is prove ##a^y(a^{x-y}-1) \geq b^y(b^{x-y}-1)##. Note each side is multiplying two non negative numbers, so it suffices to prove each piece of the multiplication is larger. Since ##a>b##, ##y\geq 0## and ##x-y\geq 0##, we get ##a^y>b^y##, and ##a^{x-y} \geq b^{x-y}##. Subtracting one from the last inequality completes the proof.
 
Office_Shredder said:
I'm pretty skeptical that you can prove Catalan's conjecture with this
I wasn't actually trying to prove it I was just playing around. Thanks for your proof by the way!
 
I noticed the question before noticing the nice answers, so already had this alternative answer:

a^b +b^a < a^a + b^b is equivalent to a^b - a^a < b^b - b^a. By the fundamental theorem of calculus, the left side is the area under the graph of y = e^x, between the points a.Ln(a) and b.Ln(a) an interval of length Ln(a).(b-a).

Similarly the right side is the area under the same graph between the points a.Ln(b) and b.Ln(b) an interval of length Ln(b).(b-a).

Assuming 0 < a < b, the second interval begins further to the right, and is also longer than the first interval,
so the amount of area is greater for the second interval. This holds for any increasing graph.

In this case, moving an interval to the right by an amount t>0, and keeping it the same length, increases the area by a factor of e^t > 1. and then making the interval also longer by moving the right endpoint further to the right increases the area more.
 
Insights auto threads is broken atm, so I'm manually creating these for new Insight articles. In Dirac’s Principles of Quantum Mechanics published in 1930 he introduced a “convenient notation” he referred to as a “delta function” which he treated as a continuum analog to the discrete Kronecker delta. The Kronecker delta is simply the indexed components of the identity operator in matrix algebra Source: https://www.physicsforums.com/insights/what-exactly-is-diracs-delta-function/ by...
Fermat's Last Theorem has long been one of the most famous mathematical problems, and is now one of the most famous theorems. It simply states that the equation $$ a^n+b^n=c^n $$ has no solutions with positive integers if ##n>2.## It was named after Pierre de Fermat (1607-1665). The problem itself stems from the book Arithmetica by Diophantus of Alexandria. It gained popularity because Fermat noted in his copy "Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos, et...
I'm interested to know whether the equation $$1 = 2 - \frac{1}{2 - \frac{1}{2 - \cdots}}$$ is true or not. It can be shown easily that if the continued fraction converges, it cannot converge to anything else than 1. It seems that if the continued fraction converges, the convergence is very slow. The apparent slowness of the convergence makes it difficult to estimate the presence of true convergence numerically. At the moment I don't know whether this converges or not.
Back
Top