Proof of the Erdos-Szekeres Bound
We prove the classical Erdos-Szekeres upper bound using an elegant combinatorial argument that combines induction with Pascal's identity.
Statement
For all integers , In particular, .
Proof
We proceed by induction on . The base cases are for all , and .
For the inductive step, assume and that and .
We first establish the recursive inequality:
Let and consider any 2-coloring of the edges of with colors red and blue. Fix a vertex and partition the remaining vertices into:
Since , either or .
If , then the coloring restricted to contains either a red (which with gives a red ) or a blue . If , then the coloring restricted to contains either a red or a blue (which with gives a blue ).
Now apply the inductive hypothesis: where the last step is Pascal's identity.
Asymptotic Analysis
Setting gives By Stirling's approximation, , so .
Erdos proved using the probabilistic method (1947). The gap between and has been one of the most famous open problems in combinatorics. In 2023, Campos, Griffiths, Morris, and Sahasrabudhe achieved the first exponential improvement to the upper bound, showing for some explicit .
For , much tighter bounds are known: . The upper bound follows from Shearer's entropy method and the lower bound from the triangle-free process (Bohman-Keevash, Fiz Pontiveros-Griffiths-Morris).