Mathematical Tripos Part III Paper 12: Probabilistic Combinatorics Exam Questions, Exams of Mathematics

The questions from the probabilistic combinatorics section of the mathematical tripos part iii exam held on june 7, 2010. The questions cover topics such as random hypergraphs, concentration of probability inequalities, and talagrand distance. Students are required to attempt no more than three questions out of the four provided, and the questions carry equal weight.

Typology: Exams

2012/2013

Uploaded on 02/26/2013

devanan
devanan 🇮🇳

4.5

(2)

75 documents

1 / 3

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
MATHEMATICAL TRIPOS Part III
Monday, 7 June, 2010 9:00 am to 11:00 am
PAPER 12
PROBABILISTIC COMBINATORICS
Attempt no more than THREE questions.
There are FOUR questions in total.
The questions carry equal weight.
STATIONERY REQUIREMENTS SPECIAL REQUIREMENTS
Cover sheet None
Treasury Tag
Script paper
You may not start to read the questions
printed on the subsequent pages until
instructed to do so by the Invigilator.
pf3

Partial preview of the text

Download Mathematical Tripos Part III Paper 12: Probabilistic Combinatorics Exam Questions and more Exams Mathematics in PDF only on Docsity!

MATHEMATICAL TRIPOS Part III

Monday, 7 June, 2010 9:00 am to 11:00 am

PAPER 12

PROBABILISTIC COMBINATORICS

Attempt no more than THREE questions. There are FOUR questions in total. The questions carry equal weight.

STATIONERY REQUIREMENTS SPECIAL REQUIREMENTS

Cover sheet None Treasury Tag Script paper

You may not start to read the questions printed on the subsequent pages until instructed to do so by the Invigilator.

2

1

(i) Let G (^) N(r,p) be the random r-uniform hypergraph with vertex set N in which every r-set

is chosen with probability p, where 0 < p < 1. Show that a.s. G (^) N(r,p) is isomorphic to a

fixed r-uniform hypergraph G (^) univ(r).

(ii) Delete the maximal vertex of each hyperedge E of G (^) univ(3) from E to obtain a graph G 0.

Show that G 0 is isomorphic to G (^) univ(2).

(i) Prove a concentration of probability inequality for a Lipschitz function on the symmetric group Sn, considered as a probability space, with Hamming distance.

(ii) Let A and B be non-empty subsets of Sn, and write d(A, B) for their Hamming distance. Show that min{|A|, |B|} 6 n! e−d (^2) / 8 n .

[The results you use should be clearly stated.]

Let z = (z 1 ,... , zn) be a sequence of n points chosen at random in the right-angled triangle with vertices (0, 0), (1, 0) and (1, 1), and let Ln = Ln(z) be the maximal number of points zi forming a convex polygon with (0, 0) and (1, 1). [Thus Ln(z) is the maximal ℓ such that the points (0, 0), zi 1 , zi 2 ,... , ziℓ and (1, 1) are the vertices of a convex (ℓ + 2)- gon.]

(i) Sketch a proof of the fact that the median of the random variable Ln is at most cn^1 /^3 for some c > 0. [You may find it helpful to consider the triangles formed by the tangents of the hyperbola y = x^2 at the points (k/n^1 /^3 , k^2 /n^2 /^3 ).]

(ii) Let 1 6 b < a be positive integers, and set A = {z : Ln(z) > a} and B = {z : Ln(z) 6 b}. Show that

dT (A, B) >

a − b √ a

where dT (A, B) is the Talagrand distance of the sets A and B.

(iii) Let ω(n) → ∞. Prove that whp Ln is concentrated in an interval of length ω(n)n^1 /^6.

Part III, Paper 12