Proof that algebraic numbers are countable

  • Thread starter Thread starter Mr Davis 97
  • Start date Start date
  • Tags Tags
    Numbers Proof
Click For Summary

Homework Help Overview

The discussion revolves around the proof that algebraic numbers are countable, focusing on the properties of polynomials with integer coefficients and their roots. The original poster presents a scenario involving polynomials of a specific degree and constraints on their coefficients.

Discussion Character

  • Conceptual clarification, Assumption checking

Approaches and Questions Raised

  • Participants explore the implications of allowing the leading coefficient of a polynomial to be zero and whether this affects the classification of polynomials by degree. Questions arise about the definitions of sets of roots and whether they should include roots from polynomials of lower degrees.

Discussion Status

Participants are actively questioning the assumptions made in the proof regarding polynomial degrees and the nature of coefficients. Suggestions for clarifying the proof's language and structure have been made, particularly concerning the necessity of specifying that the leading coefficient is non-zero.

Contextual Notes

There is a noted concern that the proof does not explicitly state that the coefficients must be integers, which could lead to confusion. Participants are considering how to phrase the initial conditions more clearly to avoid ambiguity.

Mr Davis 97
Messages
1,461
Reaction score
44

Homework Statement


Fix ##n,m \in \mathbb{N}##. The set of polynomials of the form ##a_nx^n + a_{n-1}x^{n-1} + \cdots + a_1x + a_0##
satisfying ##|a_n| + |a_{n-1}| + \cdots + |a_0| \le m## is finite because there are only a finite
number of choices for each of the coefficients (given that they must be integers.)
If we let ##A_{nm}## be the set of all the roots of polynomials of this form, then because
each one of these polynomials has at most ##n## roots, the set ##A_{nm}## is finite. Thus
##A_n##, the set of algebraic numbers obtained as roots of any polynomial (with
integer coefficients) of degree ##n##, can be written as a countable union of finite
sets: ##\displaystyle A_n = \bigcup_{m=1}^{\infty}A_{nm}##. It follows that ##A_n## is countable. Since the countable union of countable sets is countable, ##\displaystyle A = \bigcup_{n=1}^{\infty}A_{n}##, which is the set of all algebraic numbers, is countable.

Homework Equations

The Attempt at a Solution


I generally understand this proof, but I just have a few questions. In the first part of the proof, is it okay if ##a_n = 0##? That is, do the roots contained in ##A_{nm}## necessarily have to come from an nth degree polynomial, or can they also come from polynomials of degree n or less? For example, is ##A_{12}## a subset of ##A_{22}##?

If this is not the case, should we stipulate in the beginning that ##a_n \ne 0## so that the roots can actually be of an nth degree polynomial?
 
Last edited:
Physics news on Phys.org
Generally what is meant by saying that a polynomial has degree n is that the position of the highest-position, non-zero coefficient is n. Thus, by definition a n-th degree polynomial cannot have ##a_n=0##. In order not to disrupt this, the degree -1 is awarded to the zero polynomial (although that doesn't affect this case, as only polynomials of positive degree are used in the definition of the algebraic numbers). So ##A_{12}## is generally not a subset of ##A_{22}##.
 
  • Like
Likes   Reactions: Mr Davis 97
andrewkirk said:
Generally what is meant by saying that a polynomial has degree n is that the position of the highest-position, non-zero coefficient is n. Thus, by definition a n-th degree polynomial cannot have ##a_n=0##. In order not to disrupt this, the degree -1 is awarded to the zero polynomial (although that doesn't affect this case, as only polynomials of positive degree are used in the definition of the algebraic numbers). So ##A_{12}## is generally not a subset of ##A_{22}##.
So is it implicit in the following statement that ##a_n \ne 0##, even though ##a_n=0## could possibly satisfy the inequality stated?

The set of polynomials of the form ##a_nx^n + a_{n-1}x^{n-1} + \cdots + a_1x + a_0## satisfying ##|a_n| + |a_{n-1}| + \cdots + |a_0| \le m##
 
This depends alone on whether you consider the ##A_{n}## as all roots of polynomials of degree exactly ##n## or of degree at most ##n##. One is a graduation and the other one a filtration. The "exactly of degree ##n##" makes probably more sense here to avoid multiple counts, which means ##a_n \neq 0##.
 
  • Like
Likes   Reactions: Mr Davis 97
So to make the proof better, or at least more obvious, should I write at the beginning:

The set of polynomials of the form ##a_nx^n + a_{n-1}x^{n-1} + \cdots + a_1x + a_0## satisfying ##|a_n| + |a_{n-1}| + \cdots + |a_0| \le m##, where ##a_n \ne 0## is finite because there are only a finite number of choices for each of the coefficients (given that they must be integers.)?
 
I would do so, because this way you count all roots of polynomials of degree ##n##. To be exact you get an upper bound, as there are still double counts as for ##(x-a)^n##, but each one has at most ##n## different roots. If ##a_n=0## would be allowed, then you count all roots of polynomials of degree lesser than ##n## again and again.
 
I noticed that the proof doesn't state anywhere directly that the coefficients must be integers. Rather, it makes a comment in parentheses that implies the statement has already been made, which it hasn't.

Another way to write the first piece, taking care of that issue and maybe shortening the expression a bit, be:

For positive integers ##m,n##, let ##P_{nm}## be the set of all polynomials of degree ##n## with integral coefficients, such that the sum of the absolute values of each polynomial's coefficients is less than ##m##, and let ##A_{nm}## be the set of all roots of polynomials in ##A_{nm}##.
 

Similar threads

  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K
Replies
1
Views
2K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 24 ·
Replies
24
Views
4K
  • · Replies 3 ·
Replies
3
Views
1K