Ross Program 2025: Application Problems

This document is one part of the application to the Ross Mathematics Program. The entire
application can be found at

https://apply.rossprogram.org/

The deadline for applications is March 15, 2025. The Admissions Committee will start
reading applications after that date.
The Ross Mathematics Program values four qualities in a student above all else. These
qualities are exploration, curiosity, explanation, and rigor. The following questions are meant
to help you develop these aspects. You are not expected to answer every question perfectly;
rather, take this application as an opportunity to explore some beautiful mathematics! We
are interested in seeing how you approach unfamiliar open-ended math problems, and we
encourage you to write up whatever you discover and think about, especially conjectures or
variations of the questions, even if you can’t prove them. The Program believes that the
most valuable part of a problem is the time spent thinking about it, and your application
should reflect this.
Submit your own work on these problems. If you’ve seen one of the problems before
(e.g. in a class, online), please include a reference along with your solution. If you do use
any resources, be sure to credit them.
We are not looking for quick answers written in minimal space. Instead, we hope to see
evidence of your explorations, conjectures, proofs, and generalizations written in a readable
format.
Problem 1: Number Wheels
An n-wheel is a regular 2n-gon with a number placed at
each vertex. That wheel is “balanced” if the sum of any 5
two adjacent numbers equals the sum of the two numbers
on the opposite side of the polygon. The figure provides a 3 1
balanced 3-wheel.

(a) Find all of the balanced 2-wheels.

4 6
(b) Let’s call an n-wheel primitive if each number equals
the number on the opposite side of the polygon. 2
Verify that every primitive n-wheel is balanced.

(c) For which n does a non-primitive balanced n-wheel

exist?
Explain your reasoning.

(d) An n-wheel is classic if the 2n numbers are a rear-

rangement of the numbers 1, 2, 3, . . . , 2n.
For instance, the 3-wheel displayed above is classic.
For which n does a classic balanced n-wheel exist?

(e) Explore n-wheels. Invent your own questions.

What progress can you make on your questions?
Problem 2: Fibonacci fill-ins
Define an f -sequence to be a list of numbers (an ) = (a0 , a1 , a2 , . . . ) that satisfies the recur-
rence
an = an−1 + an−2 forn ≥ 2.

The classic Fibonacci sequence (Fn ) = (0, 1, 1, 2, 3, 5, 8, 13, . . . ) is the f -sequence obtained
by setting F0 = 0 and F1 = 1.
If we are given two consecutive terms in an f -sequence (an ), then all terms ak can be
computed using that recurrence.

(a) If a0 and a1 are given, explain why

an = Fn−1 · a0 + Fn · a1

for every n ≥ 1.

(b) If a0 = a and a4 = b, find the terms a1 , a2 , a3 in terms of a and b.

For instance, if a0 = 1 and a4 = 7, then (an ) = 1, 53 , 38 , 13


3
, 7, . . . .

(c) What property must the integers a, b satisfy to ensure that when a0 = a and a4 = b,
all entries of the f -sequence (an ) are integers?

(d) More generally, for a positive index k, what property must integers a and b have to
ensure that when a0 = a and ak = b, all entries of the f -sequence (an ) are integers?

(e) What other questions do “Fibonacci fill-ins” inspire? Can you make progress on those
questions?
Problem 3: Antiletters
Words are formed out of letters, but in the theory of antiletters, repeating letters can be
deleted or introduced to yield the “same” word. So the word ABBC equals AC.

(a) Consider an alphabet of three letters A, B, and C, and form words subject to the
rules that subwords matching AA, BB, CC, ABAB, BCBC, and CACA can be deleted or
introduced. How many distinct words can you make?
For example, ABCA = ABCACC = ABCACAAC = ABAC = ABABBC = BC, so ABCA
and BC are the same word.

(b) Now consider an alphabet of four letters A, B, C, and D, and form words where subwords
matching any of

AA, BB, CC, DD, ABAB, BCBC, CDCD, DADA

can be deleted or introduced. How many distinct words can you make?

(c) Continue the pattern, now studying words built from five letters A, B, C, D, E, where
subwords matching any of

AA, BB, CC, DD, EE, ABAB, BCBC, CDCD, DEDE, EAEA

can be deleted or introduced to yield the “same” word. Is there a sense in which five
letters results in “more” possible words than four letters?

(d) Riff on these ideas. What happens if you invent your own patterns of subwords that
can be deleted?
Problem 4: Hilbert numbers
The Hilbert numbers are the integers of the form 4k + 1, for k = 0, 1, 2, 3, . . . . A Hilbert
number is prime if it is larger than 1 and not a product of two smaller Hilbert numbers. For
example, check that 21, 33, 49, and 77 are Hilbert primes.
Hilbert integers do not factor uniquely into Hilbert primes, as shown by the example

77 × 21 = 33 × 49.

Let m be a fixed positive integer. The m-Hilbert numbers are the integers of the form mk +1,
for k = 0, 1, 2, 3, . . . , so the Hilbert numbers discussed above correspond to m = 4. Following
this analogy, we similarly define m-Hilbert primes.

(a) For which values of m do you think the m-Hilbert integers factor uniquely into m-
Hilbert primes?

(b) Neither the 4-Hilbert numbers (i.e., the Hilbert numbers) nor the 5-Hilbert numbers
factor uniquely into their corresponding primes. Can you think of a sense in which
the 4-Hilbert numbers are “closer” to having unique factorization than the 5-Hilbert
numbers? Is there an m yielding even “worse” factorization than m = 5?

(c) What variations on these questions might you ask? Share some of your progress.