4

To tell the truth, I'm not even sure what this means.

The professor gave an example saying that $\mathbb{Z}_m=\{[0],[1],[2],\dots,[m-1]\}$, and I sort of understand that.. but I have no idea what this means.

Well, I have maybe some idea, but I don't know how to determine if it's well defined or not. It looks like the function defines something to do with mod 7, but don't ask me to elaborate beyond that.


Edit:

OK, so my current understanding of the problem is as such:

$\mathbb{Z}_{30}$ is a set of 30 equivalence classes such that 0~30, 1~31, 2~32, etc... It breaks up the integers into groups such that $x\equiv x+30\ (mod\ 30)$.

$\mathbb{Z}_{30}=\{[x]\mid a\in\mathbb{Z},\ x\sim a\iff x\equiv a\ (mod\ 30)\}$

$f([a])=[7a]$ is eluding me a bit though... Is seems like it's saying that when you give $f$ an equivalence class, it transforms it to the equivalence class $7$ times the one given.. so $f([2])=[14]$, $f([29])=[203]$... But what do I do with that? $203\ mod\ 30=23$, which is not the same equivalence class as $[29]$.

Mirrana
  • 9,009
  • 1
  • 1
    Strangely enough, I was just reading about well defined functions on Gowers' blog: http://gowers.wordpress.com/2009/06/08/why-arent-all-functions-well-defined/ which does a pretty good job of explaining "why" we need to show a "function" is well defined. – Tyler Mar 07 '13 at 18:44
  • 1
    $\mathbb{Z}{30}$ is the group of integers $0$ through $29$ under the opration of addition modulo $30$. Imagine a clock face with no minute hand, $0$ in the 12:00 position and $1$ through $29$ proceeding clockwise around the clockface back to $0$ again. Then it's just like telling time. In $\mathbb{Z}{30}$, $28+4=2$. I'm not sure what the brackets mean in your question, but if we look at $f(a)=7a$ in $\mathbb{Z}_{30}$, we get stuff like $f(3)=21$, $f(5)=5$, and $f(7)=19$. Combined with the links above that might help get a grip on the problem. – Todd Wilcox Mar 07 '13 at 19:07
  • 2
    @ToddWilcox Given an equivalence relation $\textbf{~}$ on a set $X$, it's common to write $[x\textbf{]}_\textbf{~}:={a\in X: x\textbf{~}a}$. When there's no confusion about what's the equivalence relation at hand, it's usual to remove the subscript. – Git Gud Mar 07 '13 at 19:11
  • That was my guess. Agent154, I hope that concept makes sense: $\mathbb{Z}_{30}$ partitions the integers into $30$ equivalence classes, so the question is what happens to those classes under $f$? – Todd Wilcox Mar 07 '13 at 19:15
  • @ToddWilcox That's about right... I know what equivalence classes are, but I have no idea how to interpret the function to decide if it's well defined or not. – Mirrana Mar 07 '13 at 20:26

3 Answers3

6

For a function to be well-defined, you need it to be defined on each element in the domain and to give a unique element in the range as its result. Your function will be on the equivalence classes $\pmod {30}$ and the result will be an equivalence class, perhaps not the same one. The threat to being well-defined would be if it took two representatives of the same class to representatives of different classes. To take your example, you would think $f([29])=[23]$. But $29, 59, 89,$ and $329$ are all in $[29]$. If you multiply them all by $7$ and take the result $\pmod {30}$ you will get $23$, so we haven't found a counterexample.

To prove it, let $[n]$ be one of our equivalence classes. An arbitrary element of this class is $n+30k$ for $k$ some integer. Then that element is sent to $7n+210k$, which is in the same class as $7n$, so it doesn't matter which representative you take, you get the same result.

Ross Millikan
  • 374,822
  • Is this understanding (and notation) right? $\mathbb{Z}_{30}={[x]\mid \forall a\in\mathbb{Z},\ x\equiv a\ (mod\ 30)}={[0],[1],[2],\dots,[29]}$ – Mirrana Mar 07 '13 at 21:01
  • @agent154: I think the understanding is right, but no $x$ is equivalent to all the $a$'s in $\mathbb Z$. Intuitively $[x]$ is the set of all integers congruent to $x \pmod {30}$, so ${\ldots,-60+x,-30+x,x,30+x,60+x,\ldots}$. I'm not sure how to write it nicely like you tried in the center. – Ross Millikan Mar 07 '13 at 21:08
  • OK, so to check my understanding of the function - if two of the same classes - say $[29]$ and $[-1]$ are put into the function, we get out $[-7]$ and $[203]$. $[203]=[23]=[-7]$, so this is an example of it working... but it's not a counter-example needed to say it's not well defined. But in general it is well defined, because $[n]=[n+30k]$, and $[7n]=[7(n+30k)]$? – Mirrana Mar 07 '13 at 21:32
  • @agent154: The idea is right. But the class is $[29]$-you have chosen to use that as the name of the class. $-1$ and $29$ are two members of the class. The idea is that whatever member of the class you multiply by $7$, you get some member of $[23]$. – Ross Millikan Mar 07 '13 at 21:37
  • The point is that $-1 \ne 29$ is not important. What is important is that $[-1] = [29]$. – Steven Alexis Gregory Nov 27 '20 at 16:19
1

If you consider $\mathbb Z_{30}$ as a ring, you want to prove that the ring structure is maintained. For example, is $f(2\cdot 3 + 8)\equiv f(2) \cdot f(3) + f(8) \pmod {30}?$ Let's try this one: $f(2 \cdot 3 + 8)=f(14)=98 \equiv 8 \pmod {30}.$ But $f(2) \cdot f(3) + f(8)=14\cdot 21+56=350 \equiv 20 \pmod {30}$. Failure. If you consider just the group structure of $\mathbb Z_{30}$ you will succeed.

Ross Millikan
  • 374,822
  • But I have no idea what a "ring" is.. This is the first time I've seen this notation. – Mirrana Mar 07 '13 at 20:23
  • @agent154: do you know what a group is? A ring is the next step up in abstract algebra. Roughly, a ring is like the integers: you can add, subtract, and multiply with multiplication distributing over addition properly. You have identities for addition and multiplication. The integers modulo any integer (except 0 and maybe 1) form a ring, but you can have zero divisors: $ab=0$ with $a\ne 0, b \ne 0$ – Ross Millikan Mar 07 '13 at 20:29
  • My best guess at what a "group" is is an ordered set. I'm only doing intro discrete math at the moment. Abstract Algebra is next year in my university career, I think. – Mirrana Mar 07 '13 at 20:30
  • @agent154: Then I suspect you are just supposed to find that this $f$ is a bijection on the set of equivalence classes. You can just try it for all $30$ elements and find that the results are all distinct. FYI, a group is a set with an associative operation and inverses. They are very useful for study of symmetry, among other things. – Ross Millikan Mar 07 '13 at 20:41
  • @Ross, I’m not sure that your answer is apposite. OP doesn’t seem to have claimed or thought that his map $f$ was any kind of morphism. – Lubin Mar 07 '13 at 22:17
  • @Lubin: true. That is why I started the other answer. I left this one as it might be helpful to others. – Ross Millikan Mar 07 '13 at 22:44
0

First we need to understand what $\mathbb Z_{30}$ is. As your professor said $\mathbb Z_{30}=\{[0],[1],[2],\ldots,[29]\}$ is a set containing $30$ elements.
The formal definition of this set is as equivalence classes (the congruence classes) of an equivalence relation (the Congruence relation) on $\mathbb Z$. The elements of $\mathbb Z_{30}$ are actually sets.
The element $[i]\in\mathbb Z_{30}$ is the subset of $\mathbb Z$ containing all integers which leave remainder $i$ when they are divided by $30$. For example: $\{-120027,-57,3,33,63,370293\}\subseteq[3]$ since all elements of this set leave remainder $3$ when they are divided by $30$. The number $125\not\in[3]$ because the remainder of $125$ divided by $30$ is five and not three; therefore $125\in[5]$.
What is $[0]$? It is the subset of $\mathbb Z$ containing all integers with no remainder when they are divided by $30$ (i.e. they are divided evenly by $30$).
The elements of $\mathbb Z_{30}$ look like this: \begin{split} [0]&=\{\ldots,-60,-30,0,30,60,90,\ldots\},\\ [1]&=\{\ldots,-59,-29,1,31,61,91,\ldots\},\\ [2]&=\{\ldots,-58,-28,2,32,62,92,\ldots\},\\ \vdots \ & \hspace{19ex} \vdots\\ [29]&=\{\ldots,-31,-1,29,59,89,119,\ldots\}.\\ \end{split} Sometimes we use different names for the same elements of $\mathbb Z_{30}$. For example $[1]=[31]=[-29]$ and in general $[i]=[k]\iff k\in[i]$ for all $0\leq i\leq29$.

In general $\mathbb Z_n=\{[0]_n,[1]_n,\ldots,[n-1]_n\}$ (when is necessary we use the subscripts to avoid confusion) is the set containing $n-1$ elements, each element being a subset of $\mathbb Z$. The element $[i]_n\in\mathbb Z_n$ is $$[i]_n=\{k\in\mathbb Z: \text{the remainder of $k$ when is divided by $n$ is $i$}\}=\{rn+i:r\in\mathbb Z\}.$$ Another example is $\mathbb Z_2=\{[0],[1]\}$ with $[0]$ being the subset of $\mathbb Z$ containing all integers divided evenly by $2$, i.e. all even integers and $[1]$ is the rest of $\mathbb Z$, the odd integers. Therefore \begin{split} [0]_2&=\{\ldots,-46,\ldots,-24,\ldots,-4,-2,0,2,4,\ldots,36,\ldots\},\\ [1]_2&=\{\ldots,-45,\ldots,-23,\ldots,-3,-1,1,3,5,\ldots,37,\ldots\}. \end{split} Now a function $f:\mathbb Z_n \longrightarrow \mathbb Z_m$ between these sets is a relation which takes as input an element (a set) $[i]_n\in\mathbb Z_n$ and give as output an element of $\mathbb Z_m$.

What does it mean "$f:\mathbb Z_n\longrightarrow\mathbb Z_n$ defined as f([a])=[3a]"? It means $f$ maps the set represented by $[a]$ to the set represented by $[3a]$.

When we ask if such $f:\mathbb Z_n \longrightarrow \mathbb Z_m$ is well defined actually we ask: If we take different representatives $[a],[b]$ of the same set, then $f([a])=f([b])$? Since $[a]=[b]$ is the same thing then for $f$ to be well defined (to be function) we must have $f([a])=f([b])$.

Here are some examples of well defined and not well defined functions.

  • $f:\mathbb Z_n \longrightarrow \mathbb Z_n$ with $f([a])=[a]$.
    Is well defined because if $[a]=[b]\in\mathbb Z_n$ then $f([a])=[a]=[b]=f([b])$.
  • $f:\mathbb Z_2 \longrightarrow \mathbb Z_2$ defined as follows $f([a])=[x]$ where $x=r+1$ with $r$ the last digit of $a$.
    So $ \ f([34])=[5], \ f([-13])=[4], \ f([-1])=[2]$ etc. Is $f$ well defined? Lets see. The set $\mathbb Z_2$ contains $2$ elements $[0]$ and $[1]$. If $[a]=[b]$ then we have two possibilities. Either $[a]=[b]=[0]$, so $a,b$ are both even, or $[a]=[b]=[1]$ and $a,b$ are both odd integers. If $a,b$ are even then their last digit is one of $\{0,2,4,6,8\}$, therefore if $f([a])=[x]$ and $f([b])=[y]$ then $x,y\in\{1,3,5,7,9\}$ so $[x]=[y]=[1]$. Equivalently $f([a])=f([b])=[1]$. In a similar way if $a,b$ are odd $f([a])=f([b])=[0]$.
    It follows that $f$ is well defined.
  • $f:\mathbb Z_3 \longrightarrow \mathbb Z_3$ defined as follows $f([a])=[x]$ where $x=r+1$ with $r$ the last digit of $a$.
    In this case $f$ is not well defined: $[1]=[13]$ and $f([1])=[2], \ f([13])=[4]\neq[2]$.
  • $f:\mathbb Z_{47} \longrightarrow \mathbb Z_{47}$ defined as follows $f([a])=[3a]$.
    Let's say that $[a]=[b]\in \mathbb Z_{47}$. This means that $a$ and $b$ leave the same remainder when they are divided by $47$. Therefore $a=47k+i$ and $b=47m+i$ for some $k,m\in\mathbb Z$ and $0\leq i\leq46$. We want to check whether $f([a])=f([b])$; equivalently whether $[3a]=[3b]$. Since $a=47\cdot3k+3i$ and $b=47\cdot3m+3i$ it follows that the remainder of $a$ divided by $47$ is the same as the remainder of $b$ divided by $47$. Both are the remainder of $3i$ divided by $47$. Therefore $f([a])=[3a]=[3b]=f([b])$ and $f$ is well defined.
  • $f:\mathbb Z_{30} \longrightarrow \mathbb Z_{30}$ defined as follows $f([a])=[7a]$...
P..
  • 14,929