Skip to content

Latest commit

 

History

History
611 lines (411 loc) · 20.8 KB

File metadata and controls

611 lines (411 loc) · 20.8 KB
title author header-includes abstract
Abstract Algebra by Pinter, Chapter 24
Amir Taaki
- \usepackage{mathrsfs} - \usepackage{mathtools} - \usepackage{extpfeil} - \DeclareMathOperator\ker{ker} - \DeclareMathOperator\ord{ord} - \DeclareMathOperator\gcd{gcd} - \DeclareMathOperator\lcm{lcm} - \DeclareMathOperator\char{char} - \DeclareMathOperator\max{max} - \DeclareMathOperator\ran{ran} - \DeclareMathOperator\deg{deg} - \newcommand{\mod}[1]{\ (\mathrm{mod}\ #1)} - \newcommand{\repr}[1]{\overline{#1}} - \newcommand{\leg}[2]{\left(\frac{#1}{#2}\right)}
Chapter 24 on Rings of Polynomials

A. Elementary Computation in Domains of Polynomials

Q1

$\mathbb{Z}[x]$

$$a(x) + b(x) = x^3 + 7x^2 + 4x + 1$$ $$a(x) - b(x) = -x^3 - 3x^2 + 2x + 1$$ \begin{align*} a(x)b(x) &= 2x^5 + 10x^4 + 2x^3 + 3x^4 + 15x^3 + 3x^2 + x^3 + 5x^2 + x \ &= 2x^5 + 13x^4 + 18x^3 + 8x^2 + x \end{align*}

$\mathbb{Z}_5[x]$

$$a(x) + b(x) = x^3 + 2x^2 + 4x + 1$$ $$a(x) - b(x) = 4x^3 + 2x^2 + 2x + 1$$ $$a(x)b(x) = 2x^5 + 3x^4 + 3x^3 + 3x^2 + x$$

$\mathbb{Z}_6[x]$

$$a(x) + b(x) = x^3 + x^2 + 4x + 1$$ $$a(x) - b(x) = 5x^3 + 3x^2 + 2x + 1$$ $$a(x)b(x) = 2x^5 + x^4 + 2x^2 + x$$

$\mathbb{Z}_7[x]$

$$a(x) + b(x) = x^3 + 4x + 1$$ $$a(x) - b(x) = 6x^3 + 4x^2 + 2x + 1$$ $$a(x)b(x) = 2x^5 + 6x^4 + 4x^3 + x^2 + x$$

Q2

$$\mathbb{Z}: x^3 + x^2 + x + 1 = (x^2 + 3x + 1)(x - 2) + (5x - 5)$$ $$\mathbb{Z}_5: x^3 + x^2 + x + 1 = (x^2 + 3x + 2)(x + 3)$$

Q3

$$\mathbb{Z}: x^3 + 2 = (\frac{x}{2} - \frac{3}{4})(2x^2 + 3x + 4) + (\frac{x}{4} + 5)$$ $$\mathbb{Z}_3: x^3 + 2 = (2x)(2x^2 + 3x + 4) + (-2x + 2)$$ $$\mathbb{Z}_5: x^3 + 2 = (3x + 3)(2x^2 + 3x + 4) + 4x$$

Q4

a

When $n = 1$, $x + 1$ is a factor of $x^n + 1$.

Assume $n = k$ is true \begin{align*} x^{k + 2} + 1 &= x^2 x^k \ &= x^2(x^k + 1) + (1 - x^2) \ &= x^2 (x^k + 1) (1 - x)(1 + x) \end{align*}

Since $x + 1$ is a factor of $x^k + 1$, this means $x + 1$ is also a factor of $x^{k + 2}$.

b

As before $n = 1$ is trivially true and we assume $n = k$ is true.

$$x^{k + 2} + x^{k + 1} + x^k + \cdots + x + 1 = x^2(x^k + \cdots + x + 1) + (x + 1)$$

Since $x + 1$ divides both terms, that means it is a divisor of the expression on the left.

Q5

By induction assume $m = k$ is true, then $$x^{k + 1} + 2 = x(x^k + 2) + (x + 2)$$ $(x + 2)$ divides both sides and so is a divisor of $x^{k + 1} + 2$ in $\mathbb{Z}_3[x]$.

Likewise for $\mathbb{Z}_n[x]$ \begin{align*} x^{k + 1} + (n - 1) &= x(x^k + (n - 1)) + (x + (n - 1)) \ &= x^{k + 1} + (n - 1)x + x + (n - 1) \ &= x^{k + 1} + nx + (n - 1) \ &= x^{k + 1} + (n - 1) \end{align*} and so $x + (n - 1) is a factor of $x^{k + 1} + (n - 1)$ in $\mathbb{Z}_n[x]$.

Q6

\begin{align*} (2x^2 + ax + b)(3x^2 + 4x + m) &= 6x^4 + 8x^3 + 2x^2 m + 3ax^3 + 4ax^2 + max + 3bx^2 + 4bx + mb \ &= 6x^4 + 50 \end{align*} grouping terms $$6x^4 + (8 + 3a)x^3 + (2m + 4a + 3b)x^2 + (ma + 4b)x + mb = 6x^4 + 50$$

Writing out the roots, we have $$8 + 3a = 0$$ $$2m + 4a + 3b = 0$$ $$ma + 4b = 0$$ $$mb = 50$$

The first equation has no solution since $3 \nmid a$ and so $6x^4 + 50$ cannot be factored into $3x^2 + 4x + m$.

Q7

\begin{align*} (x^3 + ax^2 + bx + c)(x^2 + 1) &= x^5 + x^3 + ax^4 + ax^2 + bx^3 + bx + cx^2 + c \ &= x^5 + ax^4 + (1 + b)x^3 + (a + c)x^2 + bx + c \ &= x^5 + 5x + 6 \end{align*} Comparing terms, we have \begin{align*} a &\equiv 0 \mod n \ 1 + b &\equiv 0 \mod n \ a + c &\equiv 0 \mod n \ b &\equiv 5 \mod n \ c &\equiv 6 \mod n \ \implies 1 + 5 &\equiv 0 \mod n \ \implies 6 &\equiv 0 \mod n \end{align*}

$$n = 6, 2, 3$$

B

Q1

>>> def foo(n):
...     print((n**8 + 1)%5, (n**3 + 1)%5)
... 
>>> for i in range(5):
...     foo(i)
... 
1 1
2 2
2 4
2 3
2 0

Both sides are not equal when $x = 2, 3, 4$.

Q2

No this is impossible. If they are equal then their difference is 0.

Q3

$$ 0x^2 + 0x + 0 $$ $$ 0x^2 + 0x + 1 $$ $$ 0x^2 + 0x + 2 $$ $$ \dots $$ $$ 0x^2 + 0x + 4 $$ $$ 0x^2 + 1x + 0 $$ $$ \dots $$ $$ 0x^2 + 4x + 0 $$ $$ 1x^2 + 0x + 0 $$ $$ \dots $$ $$ 4x^2 + 4x + 4 $$

There are $5^3$ polynomials in $\mathbb{Z}_5[x]$ of degree 2 or less. There are $5^2$ polynomials in $\mathbb{Z}_5[x]$ of degree 1 or 0.

Thus there are $5^3 - 5^2$ quadratic polynomials in $\mathbb{Z}_5[x]$.

Cubic:

$$ 0x^3 + 0x^2 + 0x + 0 $$ $$ \dots $$ $$ 0x^3 + 0x^2 + 0x + 4 $$ $$ 0x^3 + 0x^2 + 1x + 0 $$ $$ \dots $$ $$ 0x^3 + 0x^2 + 4x + 4 $$ $$ 0x^3 + 1x^2 + 0x + 0 $$ $$ \dots $$ $$ 0x^3 + 4x^2 + 4x + 4 $$ $$ 1x^3 + 0x^2 + 0x + 0 $$ $$ \dots $$ $$ 4x^3 + 4x^2 + 4x + 4 $$

Answer: $5^4 - 5^3$

There are $n^{m + 1} - n^m$ polynomials of degree $m$ in $\mathbb{Z}_n[x]$.

Q4

$$(x + 1)^2 = x^2 + 2x + 1 = x^2 + 1 \text{ in } A[x] \implies \char A = 2$$ $$(x + 1)^4 = x^4 + 4x^3 + 6x^2 + 4x + 1 = x^4 + 1 \text{ in } A[x] \implies \char A = \gcd(4, 6) = 2$$ $$(x + 1)^6 = x^6 + 6x^5 + 15x^4 + 20x^3 + 15x^2 + 6x + 1 = x^6 + 2x^3 + 1 \text{ in } A[x] \implies \char A = \gcd(6, 15, 20 - 2) = 3$$

Q5

$$(2x + 2)^3 = 8x^3 + 24x^2 + 24x + 8 = 0 \text{ in } \mathbb{Z}_8[x]$$ $$\implies 2x + 2 \text{ is a divisor of } 0$$

$$(1 - 4x)(1 + 4x) = 1 - 16x^2 = 1 \text{ in } \mathbb{Z}_8[x]$$ $$\implies 1 + 4x \text{ and } 1 - 4x \text{ are invertible elements}$$

Q6

For any polynomial $b(x) \in A[x], \deg b(x) \geq 0$. If $\deg b(x) = 0$ then $xb(x) = 0$ because $b(x) = 0$. Otherwise $\deg [x \cdot b(x)] = \deg x + \deg b(x) = 1 + \deg b(x) \implies \deg [x \cdot b(x)] \geq 1$.

Since $x$ is in every non-zero polynomial domain, this means there are no polynomial fields.

Q7

Take $a(x) = x \in A[x]$, then $\deg a(x) = 1$ and $\deg [(a(x))^2] = 2$. In fact $\deg[(a(x))^n] = n$ in any ring and so there is no polynomial with a nonzero term that multiplied by $x$ produces $0$. $$x(b_0 + b_1 x + \cdots + b_m x^m)$$ where $b_m \neq 0$ in the ring, then $$\deg [a(x) \cdot b(x)] = m + 1 \neq 0$$

Q8

Idempotent: $(a(x))^2 = a(x)$ Nilpotent: $(a(x))^n = 0$ for some integer $n$.

Let $a(x) = x$, then $(a(x))^2 = x^2$, so $(a(x))^2 \neq a(x)$ and $a(x)$ is not idempotent.

Also $(a(x))^n = x^n \neq 0$ and so $a(x)$ is not nilpotent.

C. Rings $A[x]$ Where $A$ Is Not an Integral Domain

Q1

An integral domain is a commutative ring with unity having no divisors of 0.

Since $A[x]$ contains the elements from $A$, then if $A$ has zero divisors, so does $A[x]$ and hence $A[x]$ is not an integral domain.

Q2

Degree 0: $2 \times 2 = 0$ in $\mathbb{Z}_4[x]$

Degree 1: $2x \cdot 2x = 0$

Degree 2: $2x^2 \cdot 2x^2 = 0$

Q3

$5x^3(2x + 1) = 0$ in $\mathbb{Z}_10[x]$ lacks the cancellation property whereas the term $5x^3 = 0$ in $\mathbb{Z}_5[x]$ and disappears.

Q4

Any polynomials where the coefficient of the leading term is a multiple of the field size.

$\mathbb{Z}_4[x]: (2x + 3)(2x + 1) = 3$

$\mathbb{Z}_6[x]: (3x + 1)(2x + 5) = 5x + 5$

$\mathbb{Z}_9[x]: (3x + 1)(3x + 4) = 6x + 4$

Q5

$$a(x) = a_0 + a_1 x + \cdots + a_n x^n$$ $$b(x) = b_0 + b_1 x + \cdots + b_m x^m$$

$$\deg a(x) = n$$ $$\deg b(x) = m$$

$$a_n, b_m \in A : a_n \neq 0, b_m \neq 0, a_n b_m = 0$$

Thus the coefficient of x^{n + m} is 0 and so $$\deg a(x) b(x) < \deg a(x) + \deg b(x)$$

Q6

In an integral domain $$\deg a(x)b(x) = \deg a(x) + \deg b(x)$$ Non-constant polynomials have a degree greater than one. Let $a(x)$ be such a polynomial, while $b(x)$ is a non-zero polynomial such that $\deg b(x) \geq 1$. Then $\deg a(x)b(x) > 1$, while the degree of 1 is 1. So there are no non-constant invertible polynomials in integral domains.

In $\mathbb{Z}_4[x], (2x + 1)^2 = 1$, so $(2x + 1)$ is invertible and so are all powers of $(2x + 1)^k$ since $(2x + 1)$ is its own inverse.

Q7

$\mathbb{Z}_9[x]$

$$(x + 3)(x + 6)$$ $$(2x + 3)(5x + 6)$$ $$(4x + 3)(7x + 6)$$ $$(5x + 3)(8x + 6)$$ $$(2x + 6)(5x + 3)$$ $$(4x + 6)(7x + 3)$$ $$(5x + 6)(8x + 3)$$

$\mathbb{Z}_5[x]$

$$5 \mid (a + b) \qquad 5 \mid ab$$ but $\gcd(a, 5) = 1$ and $\gcd(b, 5) = 1$ since $5$ is prime. So there is only 1 factorization which is $x^2$.

Q8

$\mathbb{Z}_5[x]$

\begin{align*} a + b &\equiv 1 \mod 5 \ ab &\equiv 4 \mod 5 \end{align*}

\begin{align*} 2 + 4 &\equiv 1 \mod 5 \ 3 + 3 &\equiv 1 \mod 5 \ 2 \times 2 &\equiv 4 \mod 5 \ 3 \times 3 &\equiv 4 \mod 5 \end{align*}

\begin{align*} (x + 3)^2 &= x^2 + x + 4 \ [4(x + 3)]^2 &= 16x^2 + 96x + 144 \ &= x^2 + x + 4 \ &= (4x + 2)^2 \end{align*}

$\mathbb{Z}_8[x]$

Any polynomial of the form $1 + 4x + 4x^2 + \cdots + 4x^n$ when squared will equal 1, because every coefficient apart from the constant and leading term is greater than or equal to 2, and $4\times 2 = 8 = 0$, and the leading term is $16x^{2n} = 0$. So there are infinite polynomial square roots in $\mathbb{Z}_8[x]$.

D. Domains $A[x]$ Where $A$ Has Finite Characteristic

Q1

Every coefficient in $A[x]$ is a member of $A$. For all $a(x), b(x) \in A[x], c(x) = a(x) + b(x)$ then $c_i = a_i + b_i$, and therefore the characteristic is preserved since $\underbrace{1_A + 1_A + \cdots + 1_A}_{\char A} = 0$.

Q2

Consider the ring $\mathbb{Z}_n[x]$ of polynomials in one variable $x$ with coefficients in $\mathbb{Z}_n$. It is an infinite ring since $x^m \in \mathbb{Z}_n[x]$ for all positive integers $m$, and $x^{m_1} \neq x^{m_2}$ for $m_1 \neq m_2$. But the characteristic of $\mathbb{Z}_n[x]$ is clearly $n$.

Q3

\begin{align*} (x + 2)(x^{m - 1} + x^{m - 2} + \cdots + x^2 + x + 1) &= x(x^{m - 1} + x^{m - 2} + \cdots + x^2 + x + 1) + 2(x^{m - 1} + x^{m - 2} + \cdots + x^2 + x + 1) \ &= x^m + (x^{m - 1} + x^{m - 2} + \cdots + x^3 + x^2 + x) + 2(x^{m - 1} + x^{m - 2} + \cdots + x^2 + x) + 2 \ &= x^m + 2 \end{align*}

Likewise the above applies for $(p - 1)$ in any domain of characteristic $p$.

Q4

By the cancellation property, the characteristic of every integral domain is prime, since if the characteristic was composite that would imply $rs = 0$ for some $r, s \in A$ which violates the zero divisor rule.

Thus the coefficients for all terms in the expansion $(x + c)^p$ except $x^p$ and $c^p$, by the binomial formula are equal to $\binom{p}{k} = \frac{p!}{k!(p - k)!}$. Since $p$ is prime and indivisible the coefficient becomes zero. $$(x + c)^p = x^p + c^p$$

Q5

They aren't the same since $x \notin A$, and $\forall a \in A$, $a = a^2$ but $x \neq x^2$.

Q6

It is trivial to see that \begin{align*} [a_0 + (a_1x + \cdots + a_n x^n)]^p &= a_0^p + [a_1 x + (a_2 x^2 + \cdots + a_n x^n)]^p \ &= a_0^p + a_1^p x^p + [a_2x^2 + (a_3 x^3 + \cdots + a_n x^n)]^p \ &= a_0^p + a_1^p x_1^p + \cdots + a_n^p x^{np} \end{align*}

E. Subrings and Ideals in A[x]

Q1

$B[x]$ contains all the polynomials with coefficients in $B$. Since $B$ is a subring of $A$, so $B[x]$ is a subring of $A[x]$.

Q2

Likewise $B$ absorbs all products with $A$, and hence so does $B[x]$,

Q3

Every coefficient $a_i$ with odd $i$ equal to zero, means the polynomial only has non-zero coefficients for even powers.

When adding polynomials, we add the coefficients. So the odd numbered powers remain zero, and even powers remain non-zero.

For multiplying two polynomials $a(x)b(x)$, the corresponding powers of each term are added together, $a_i b_j x^{i + j}$. Since both $i$ and $j$ are even, so is the resulting term and hence the result of $a(x)b(x)$ remains inside the set $S$ making it a subring.

The above statement does not apply when talking about odd non-zero coefficients, since multiplying two odd terms might result in an even power, for example $c(x) = a(x) b(x)$, $a_3 b_5 x^{3 + 5}$.

Q4

Let $b(x) \in A[x]$ and $a(x) \in J$, then the constant term in $b(x)$ is $b_0$. Since $b(x) a(x) = b_0 a(x) + b_1 x a(x) + \cdots + b_m x^n a(x)$, and the powers of all terms in $a(x)$ are $\geq 1$, so $b_0 a(x)$ has no constant term. So $\forall a(x) \in J$ absorbs products from $A[x]$ and is an ideal.

Q5

Let $a(x) = a_0 + a_1x + \cdots + a_n x^n \in J$ and $b(x) = b_0 + b_1x + \cdots + b_m x^m \in A[x]$. Then $a(x)b(x) = a_0 (b_0 + b_1x + \cdots + b_m x^m) + a_1x (b_0 + b_1x + \cdots + b_m x^m) + \cdots + a_n x^n (b_0 + b_1x + \cdots + b_m x^m)$. Then it can be seen plainly that the sum of cofficients for the result is $(a_0 + a_1 + \cdots + a_n)(b_0 + b_1 + \cdots + b_m) = 0$. Therefore $J$ is an ideal of $A[x]$.

Q6

Since $A$ is an integral domain, there are no divisors of zero. Therefore the values cannot be made to equal 0 unless one of the terms is zero. In the case of Q4, the polynomial is an ideal in $J$ with a zero constant coefficient and in Q5, the polynomial can be factorized into a polynomial where one of the terms has coefficients that sum to zero.

F. Homomorphisms of Domains of Polynomials

Q1

$$a(x) = a_0 + a_1 x + \cdots + a_n x^n$$ $$b(x) = b_0 + b_1 x + \cdots + b_m x^m$$

$$h(a(x) + b(x)) = h((a_0 + b_0) + \cdots) = a_0 + b_0 = h(a(x)) + h(b(x))$$ $$h(a(x)b(x)) = h(a_0 b_0 + \cdots) = a_0 b_0 = h(a(x))h(b(x))$$

Q2

$$\forall a(x) \in A[x], h(x \cdot a(x)) = h(x(a_0 + \cdots + a_n x^n)) = h(a_0 x + \cdots + a_n x^{n + 1}) = 0$$ $$\implies \ker h = { x \cdot a(x) : a(x) \in A[x] } = \langle x \rangle$$

By the definition of a principal ideal, let $x$ remain fixed as it is multiplied by elements from $A[x]$.

Q3

$$h : A[x] \rightarrow A, \ker h = \langle x \rangle \implies A[x] / \langle x \rangle \cong A$$

Q4

$$g(a(x)) = g(a_0 + \cdots + a_n x^n) = a_0 + \cdots + a_n$$

\begin{align*} g(a(x) + b(x)) &= g(a_0 + a_1 x + \cdots + a_m x^m + \cdots + a_n x^n + b_0 + b_1 x + \cdots + b_m x^m) \ &= (a_0 + b_0) + (a_1 + b_1) + \cdots + (a_m + b_m) + \cdots + a_n \ &= g(a(x)) + g(b(x)) \end{align*}

\begin{align*} g(a(x)b(x)) &= g(a_0 b(x) + a_n x^n b(x)) \ &= g(a_0 b_0 + \cdots + a_0 b_m x^m + \cdots + a_n b_0 x^n + \cdots + a_n b_m x^{n + m}) \ & = a_0 b_0 + \cdots + a_0 b_m + \cdots + a_n b_m = g(a(x)) g(b(x)) \end{align*}

Let $a \in A$, then $a(x) \in J + a$, where $J$ is the ideal of $g$ (coefficients that sum to zero). Thus every value in $A$ is an image of an element in $A[x]$ and so $h$ is surjective.

The kernel of $g$ is described in 24E5: let $J$ consist of all the polynomials $a_0 + a_1 x + \cdots + a_n x^n$ in $A[x]$ such that $a_0 + a_1 + \cdots + a_n = 0$.

Q5

\begin{align*} h(a(x) + b(x)) &= (a_0 + b_0) + (a_1 + b_1) cx + (a_2 + b_2)c^2 x^2 + \cdots + (a_n + b_n) c^n x^n \ &= (a_0 + a_1 cx + a_2 c^2 x^2 + \cdots + a_n c^n x^n) + (b_0 + b_1 cx + b_2 c^2 x^2 + \cdots + b_n c^n x^n) \ &= h(a(x)) + h(b(x)) \end{align*}

\begin{align*} h(a(x)b(x)) &= a_0 b_0 + (a_0 b_1 + a_1 b_0) cx + (a_0 b_2 + a_1 b_1 + a_0 b_2) c^2 x^2 + \cdots + \sum_{i + j = n} a_i b_j c^n x^n \ &= h(a(x))h(b(x)) \end{align*}

Since $A$ is an integral domain and there are no zero divisors, then $\ker h = { 0 }$.

Q6

Any polynomial $a(x) = a_0 + a_1 x + \cdots + a_n x^n$ can be produced by $h$ iff $c$ is invertible by setting the input to $a_0 + c^{-1} a_1 x + \cdots + c^{-n} a_n x^n$. Then the output of $h$ on this value will produce $a(x)$. Thus $h$ is an automorphism in this case.

G. Homomorphisms of Polynomial Domains Induced by a Homomorphism of the Ring of Coefficients

Q1

$$\bar{h}(a_0 + a_1 x + \cdots + a_n x^n) = h(a_0) + h(a_1)x + \cdots + h(a_n)x^n$$

\begin{align*} \bar{h}(a(x) + b(x)) &= \bar{h}((a_0 + b_0) + (a_1 + b_1)x + \cdots + (a_n + b_n) x^n) \ &= h(a_0 + b_0) + h(a_1 + b_1) x + \cdots + h(a_n + b_n) x^n \ &= (h(a_0) + h(b_0)) + (h(a_1) + h(b_1)) x + \cdots + (h(a_n) + h(b_n)) x^n \ &= \bar{h}(a(x)) + \bar{h}(b(x)) \end{align*}

\begin{align*} \bar{h}(a(x)b(x)) &= \bar{h}(a_0 b_0 + a_0 b_1 x + \cdots + a_n b_n x^{2n}) = h(a_0 b_0) + h(a_0 b_1) x + \cdots + h(a_n b_n) x^{2n} \ &= h(a_0) h(b_0) + h(a_0) h(b_1) x + \cdots + h(a_n) h(b_n) x^{2n} \ &= \bar{h}(a(x))\bar{h}(b(x)) \end{align*}

Q2

$$\forall a_i : 0 \leq i \leq n, a_i \in \ker h$$ $$a(x) = a_0 + \cdots + a_n x^n$$

Q3

If $h$ is surjective, then every element of $B$ is of the form $h(a)$ for some $a$ in $A$. Thus, any polynomial with coefficients in $B$ is of the form $h(a_0) + h(a_1)x + \cdots + h(a_n)x^n = \bar{h}(a_0 + a_1 x + \cdots + a_n x^n)$.

Q4

Every coefficient of $A[x]$ maps to a distinct coefficient in $B[x]$ because $h$ is an injective function.

Q5

$$b(x) = q(x)a(x)$$ $$\bar{h}(b(x)) = \bar{h}(q(x))\bar{h}(a(x))$$

Q6

Every coefficient $a_i = qn$ and so $h(a_i) = 0$ because $n \mid a_i$. Thus $\bar{h}(a(x)) = 0$.

Q7

$\mathbb{Z}_n$ where $n$ is prime, means the domain of $\bar{h}$ is an integral domain. $$\bar{h} : \mathbb{Z}[x] \underset{\ker \bar{h}}{\relbar\joinrel\twoheadrightarrow} \mathbb{Z}_n[x]$$

From 19F2, $J$ is a prime ideal iff $A/J$ is an integral domain. So in our case this means $\ker \bar{h}$ is a prime ideal.

An ideal $J$ of a commutative ring is said to be a prime ideal if for any two elements $a$ and $b$ in the ring, $$\text{If } ab \in J \text{ then } a \in J \text{ or } b \in J$$

$$a(x)b(x) \in \ker \bar{h} \implies a(x) \text{ or } b(x) \in \ker \bar{h}$$

H. Polynomials in Several Variables

Q1

*Prove $A$ is an integral domain $\implies A[x]$ is an integral domain.$

Given any $A_i[x_{i + 1}]$ is an integral domain, we know that the leading term $a_k \neq 0$ (which includes the other non-zero $x$ values), multiplied by another $b_l \neq 0$, and so $a_k b_l \neq 0$ and therefore $A_i[x_{i + 1}]$ has a non-zero coefficient.

Q2

Degree of $p(x, y)$ is the greatest $n$ such that the coefficient $a_n$ is non-zero for the powers $x^i y^j$ such that $i + j = n$.

$$0, 1, 2$$ $$x, x + 1, x + 2$$ $$2x, 2x + 1, 2x + 2$$ $$x^2, x^2 + 1, x^2 + 2$$ $$x^2 + x, x^2 + x + 1, x^2 + x + 2$$ $$\cdots$$ $$2x^3 + 2x^2 + 2x, 2x^3 + 2x^2 + 2x + 1, 2x^3 + 2x^2 + 2x + 2$$

Q3

\begin{align*} a(x, y) + b(x, y) &= (a_{0,0} + b_{0, 0}) + (a_{1, 0} + b_{1, 0})x + \cdots + (a_{n, 0} + b_{n, 0}) x^n \ &\quad + (a_{0, 1} + b_{0, 1}) y + (a_{1, 1} + b_{1, 1}) xy + \cdots + (a_{n, 1} + b_{n, 1}) x^n y + \cdots \ &\quad + (a_{0, n} + b_{0, n}) y^n + (a_{1, n} + b_{1, n})xy^n + \cdots + (a_{n, n} + b_{n, n})x^n y^n \ &= \sum_{i = 0}^n \sum_{j = 0}^n (a_{i,j} + b_{i, j}) x^i y^j \end{align*}

\begin{align*} a(x, y) b(x, y) &= a_{0, 0} b_{0, 0} + (a_{0, 0} b_{1, 0} + a_{0, 1} b_{0, 0}) x \ &\quad + (a_{0, 0} b_{2, 0} + a_{1, 0} b_{1, 0} + a_{2, 0} b_{0, 0}) x^2 + \cdots + a_{n, 0} b_{n, 0} x^{2n} \ &\quad + (a_{0, 1} b_{1, 0} + a_{1, 1} b_{0, 0} + a_{0, 0} b_{1, 1} + a_{1, 0} b_{0, 1}) xy \ &\quad + (a_{0, 1} b_{2, 0} + a_{0, 0} b_{2, 1} + a_{1, 1} b_{1, 0} + a_{1, 0} b_{1, 1} + a_{2, 1} b_{0, 0} + a_{2, 0} b_{0, 1}) x^2 y \ &\quad + \cdots + a_{n, n} b_{n, n} x^{2n} y^{2n} &= (c_{0, 0} + c_{1, 0} x + \cdots + c_{2n, 0} x^{2n}) \ &\quad + c_{0, 1} y + c_{1, 1} xy + \cdots + c_{2n, 1} x^{2n} y \ &\quad + \cdots + c_{0, 2n} y^{2n} + c_{1, 2n} xy^{2n} + \cdots + c_{2n, 2n} x^{2n} y^{2n} \ &= \sum_{i = 0}^{2n} \sum_{j = 0}^{2n} c_{i,j} x^i y^j \end{align*}

$$c_{k, l} = \sum_{i_x + j_x = k, ; i_y + j_y = l}{a_{i_x, i_y} b_{j_x, j_y}}$$

Q4

If there are two or more terms with the same degree, we ignore them since they do not cancel. For example $xy$ and $y^2$.

The coefficient for the leading term is of the form $$a_{m, s} b_{n, t} \text{ for } a(x, y) b(x, y)$$

Thus $\deg a(x, y) b(x, y) = (m + n) + (s + t)$ $$\deg a(x, y) b(x, y) = \deg a(x, y) + \deg b(x, y)$$

I. Fields of Polynomial Quotients

Q1

$A$ is a finite integral domain means it is a field with $\char(A)$ for $1_A$. The unity for $A(x)$ is $[1_A, 1_A]$ and $[a, b] + [c, d] = [ad + bc, bd]$. $$[1_A, 1_A] + [1_A, 1_A] = [2_A, 1_A]$$ $$[k_A, 1_A] + [1_A, 1_A] = [k_A + 1_A, 1_A]$$ \begin{align*} \underbrace{[1_A, 1_A] + \cdots + [1_A, 1_A]}_{\char(A)} &= [\char(A), 1_A] \ &= [0_A, 1_A] \end{align*}

Q2

$\mathbb{Z}_p$ is a finite field with characteristic $p$. Therefore the field of quotients $\mathbb{Z}_p(x)$ will have characteristic $p$ yet it is infinite because terms have any positive integer value (and indeed negative since $\mathbb{Z}_p$ has inverses because it is a field).

Q3

\begin{align*} \bar{h}\left(\frac{a(x)}{s(x)}\right) &= \bar{h}\left(\frac{a_0 + \cdots + a_n x^n}{s_0 + \cdots + s_n x^n}\right) \ &= \frac{h(a_0) + \cdots + h(a_n) x^n}{h(s_0) + \cdots + h(s_n)x^n} \end{align*}

Because $h$ is isomorphic, each element of $B(x)$ is the image of no more than one element of $A(x)$, so $\bar{h}$ is injective.

Likewise every element of $B(x)$ is the image of an element in $A(x)$, so $\bar{h}$ is surjective.

$\therefore \bar{h}$ is an isomorphism.

J. Division Algorithm: Uniqueness of Quotient and Remainder

In the division algorithm, prove that $q(x)$ and $r(x)$ are uniquely determined. [HINT: Suppose $a(x) = b(x)q_1 (x) + r_1 (x) = b(x)q_2 (x) + r_2 (x)$, and subtract these two expressions, which are both equal to $a(x)$.]

$$ 0 = b(x)(q_1(x) - q_2(x)) + (r_1(x) - r_2(x)) $$

Assume $\deg b(x) > 0$.

If $q_1(x) \neq q_2(x)$, then $\deg [q_1(x) - q_2(x)] > 0$ so $\deg [b(x)(q_1(x) - q_2(x)] > 0$.

But the entire expression is 0 and so its degree is zero. Hence $b(x)(q_1(x) - q_2(x))$ cannot have a degree higher than 0 so the term can only equal 0, which means $q_1(x) = q_2(x)$ since $b(x) \neq 0$.

$$\implies r_1(x) - r_2(x) = 0$$