January 19th, 2013, 03:13 AM  #1 
Member Joined: Aug 2012 Posts: 72 Thanks: 0  subgroup vs. coset
What is the different between subgroup to coset?

January 19th, 2013, 10:59 AM  #2 
Senior Member Joined: Nov 2011 Posts: 100 Thanks: 0  Re: subgroup vs. coset
Here's a reference that might help http://www.math.rochester.edu/people/fa ... pak/N3.pdf If G is a group and N<G is a normal subgroup, you can consider G/N. Elements of G/N look like gN, for g in G, and are cosets. Or you can just consider cosets of a subgroup H<G as the set gH={gh\in G: h\in H}. 
January 19th, 2013, 11:13 AM  #3  
Senior Member Joined: Aug 2012 Posts: 2,357 Thanks: 740  Re: subgroup vs. coset Quote:
In the additive group of integers, the even numbers are a subgroup. The odd numbers are one of the cosets of the even numbers; but the odd numbers are not a subgroup.  
January 19th, 2013, 11:24 PM  #4 
Senior Member Joined: Mar 2012 Posts: 294 Thanks: 88  Re: subgroup vs. coset
well, every subgroup is a coset (of itself) but not every coset is a subgroup. i like to think of it this way: imagine a group G is like a cake. suppose we have a subgroup H that's a certain size. if we cut G into "Hsized slices", the slices are the cosets, and the subgroup H is the slice that contains the identity. you can also think of it this way: suppose we have H living in G. the coset gH is H "translated" by g. it's easier to picture this if G is abelian, and we write g+H, instead of gH. so, for example, if G = RxR, the real plane, and * is componentwise addition, we have the subgroup: H = {(x,0) in R} (also known as "the xaxis"). in this case (a,b) + H is the set of all points (x,y) of the form (a+x,b), that is: the line y = b (which is parallel to H). it's not hard to see that: (a,b) + H = (0,b) + H, since (a,b)  (0,b) = (a,0) is in H. so (a,b) + H is just "the xaxis shifted by b". so the set of (left) cosets (RxR)/(Rx{0}) = G/H is the collection of all horizontal lines....and H is the one containing (0,0) (the one that goes through the origin). equivalently: you can view "multiplication by g" (where g is an element of G) as a FUNCTION: Lg(x) = g*x (here, Lg is supposed to remind you of L(eft multiplication by )g). so the coset gH is the image of H under Lg. for every g, the function Lg is bijective, because it has the inverse function Lg^1 (multplication by g^1): Lg(Lg^1(x)) = x = id(x). this means that gH is always "the same size" as H. now if g is in H, then g*h is in H (by closure), so for h in H, hH = H. but if g ISN'T in H, none of gH is in H. where this gets interesting is that we can have: gH = g'H, even when g does not equal g'. this happens when g = g'h (or viceversa), or put another way: g'^1g is in H. for groups: a quantity like a^1b roughly measures how "different" a and b are: if they are the same, we get e, the identity. if they are close (they lie within the same "translate" of H), then a^1b is in H (which you can think of as being "the closest coset to home", where "home" here means the identity of the group). 

Tags 
coset, subgroup 
Search tags for this page 
Click on a term to search for related topics.

Thread Tools  
Display Modes  

Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Can anyone help me with this coset problem?  Elladeas  Abstract Algebra  2  May 7th, 2011 08:51 AM 
Subgroup/Normal Subgroup/Automorphism Questions  envision  Abstract Algebra  3  October 4th, 2009 10:37 PM 
Subgroup/Normal Subgroup/Factor Group Questions  envision  Abstract Algebra  1  October 4th, 2009 03:24 AM 
member of coset with same order as coset  signaldoc  Abstract Algebra  3  February 19th, 2009 05:50 PM 
basic coset stuff.  mathstresser  Abstract Algebra  1  November 18th, 2007 05:14 PM 