updated:

A Computational Approach to Number Theory and Algebra - Solution


If you've found anything wrong (or have questions), you can reach me at [email protected].

Chapter 1

1.2 Ideals and greatest common divisors

Exercise 1.8

As II is already closed under addition, it's sufficient to show that ∀a∈I,−a∈I\forall a \in I, -a \in I implies II is closed under multiplication with any z∈Zz \in \mathbb{Z} and verse versa.

On one hand, the closing under addition property already implies that ∀a∈I,z∈Z+,az∈I\forall a \in I, z \in \mathbb{Z}^+, az \in I by repeatedly adding aa for zz times. And this can be easily extended to negative number z′∈Z−z' \in \mathbb{Z}^- by repeatedly adding −a-a for ∣z′∣|z'| times since −a∈I-a \in I. Simultaneously, by choosing any a∈Ia \in I, a+(−a)=0∈Ia + (-a) = 0 \in I, we can deduce that a⋅0=0∈Ia \cdot 0 = 0 \in I.

On the other hand, II is an ideal instantly implies ∀a∈I,−a∈I\forall a \in I, -a \in I by multiplying aa with −1∈Z-1 \in \mathbb{Z}.

So II is an ideal if and only if all conditions in the question holds.

Exercise 1.9

Observation 1: ∀a′>∣a∣,a′∤a\forall a' \gt |a|, a' \nmid a

Observation 2: ∀a∈Z,∣a∣∣a\forall a \in \mathbb{Z}, |a| \mid a: proof: 1⋅a=a∧(−1)⋅(−a)=a1 \cdot a = a \land (-1) \cdot (-a) = a

(a)

Suppose gcd⁡(a,b)=d\gcd(a, b) = d and gcd⁡(b,a)=d′\gcd(b, a) = d'. By definition of gcd⁡\gcd we have d∣a∧d∣bd \mid a \land d \mid b and d′∣a∧d′∣bd' \mid a \land d' \mid b, thus d∣d′∧d′∣d⇒d=d′d \mid d' \land d' \mid d \Rightarrow d = d'

(b)

gcd⁡(a,b)=∣a∣\gcd(a, b) = |a| implies ∣a∣∣b|a| \mid b, thus a∣ba \mid b. When a∣ba \mid b, we have ∣a∣∣a∧∣a∣∣b|a| \mid a \land |a| \mid b. By observation 1 we have gcd⁡(a,b)=∣a∣\gcd(a, b) = |a|

(c)

  • The first equation: Firstly, 0⋅∣a∣=0⇒∣a∣∣00 \cdot |a| = 0 \Rightarrow |a| \mid 0. Secondly, by observation 1 and 2 its easy to show that ∣a∣|a| is the maximum integer that divides aa. So gcd⁡(a,0)=gcd⁡(a,a)=∣a∣\gcd(a, 0) = \gcd(a, a) = |a|.

  • The second equation: 1∣1∧1∣a1 \mid 1 \land 1 \mid a and 11 is the only divisor of 11, thus gcd⁡(a,1)=1\gcd(a, 1) = 1

(d)

Suppose gcd⁡(ca,cb)=d\gcd(ca, cb) = d, by theorem 1.8, there exists integers s,ts, t so that sca+tcb=dsca + tcb = d, thus sa+tb=dcsa + tb = \frac{d}{c}. We can easily show that d∣c∣\frac{d}{|c|} is the minimal value for s′a+t′bs'a + t'b for integers s′,t′s', t' by proof of contradiction, thus gcd⁡(a,b)=d∣c∣\gcd(a, b) = \frac{d}{|c|}. So gcd(ca,cb)=∣c∣gcd⁡(a,b)gcd(ca, cb) = |c|\gcd(a, b).