Understanding inner product space and matrix representations of Operat

In summary: O}(f)(x)=\mathbf{O}(f)(x)+i\mathbf{O}(d/dx)(x)##This is just an informal way of representing how the operator behaves.
  • #1
cookiemnstr510510
162
14
Homework Statement
Image attached
Relevant Equations
...
(scroll to bottom for problem statement)
Hello,
I am wondering if someone could break down the problem statement in simpler terms (not so math-y).
I am struggling with understanding what is being asked.
I will try to break it down to the best of my ability

Problem statement:Consider the inner product space consisting of all linear combinations of sin(x) and cos(x).
My Understanding: To me this part is referring to the length that sin(x) and cos(x) can make with all possible linear combinations of each other. This doesn't make sense to me because inner product space seems to be talking about a literal space. After re-reading a chapter on inner product in my linear algebra book what it seemed to say was that the inner product space allows us to define things like length, angle etc. between vectors. So maybe that's what they are referring to in the problem statement.

Problem statement: what is the matrix representation of the operator i(d/dx) if we take sin(x) and cos(x) to be our basis functions?
My understanding: if sin and cos(x) are our basis then this is like saying i(hat) and j(hat) are our basis for R-2. I don't understand anything more than that.

I think Ill start with that rather than going through all parts. I want to have a good understanding of this and I am open for any suggestions. I have tried to brush up on my linear algebra, but I struggle with making connections between this and Lin Alg.

Thanks
P1.png
 
Physics news on Phys.org
  • #2
cookiemnstr510510 said:
Homework Statement:: Image attached
Relevant Equations:: ...

To me this part is referring to the length that sin(x) and cos(x) can make with all possible linear combinations of each other.
It is unclear what you mean by ”the length”. A linear combination of sin(x) an cos(x) is a function on the form f(x) = A sin(x) + B cos(x).

cookiemnstr510510 said:
Homework Statement:: Image attached
Relevant Equations:: ...

This doesn't make sense to me because inner product space seems to be talking about a literal space.
What do you mean by ”literal” space. The space being referred to is the function space consisting of all possible linear combinations of sin(x) and cos(x).

cookiemnstr510510 said:
Homework Statement:: Image attached
Relevant Equations:: ...

Problem statement: what is the matrix representation of the operator i(d/dx) if we take sin(x) and cos(x) to be our basis functions?
My understanding: if sin and cos(x) are our basis then this is like saying i(hat) and j(hat) are our basis for R-2. I don't understand anything more than that.
i(d/dx) is a (linear) operator on your function space. What do you get of you act with it on a general function in your function space?
 
  • Like
Likes cookiemnstr510510
  • #3
Bit hard to help here (guidelines require an attempt, not just 'dunno')
And you are (understandably) not very specific in where precisely you are stuck.

So let me just ask:
Can you write ##|p\rangle ## in terms of the basis functions ? ##|q\rangle ## ? ##\langle p|q \rangle## ?

Is this enough to start on b) ?
 
  • #4
BvU said:
Bit hard to help here (guidelines require an attempt, not just 'dunno')
And you are (understandably) not very specific in where precisely you are stuck.

So let me just ask:
Can you write ##|p\rangle ## in terms of the basis functions ? ##|q\rangle ## ? ##\langle p|q \rangle## ?

Is this enough to start on b) ?
The OP has made an attempt at interpreting the problem statement inside the problem statement. The problem is with this interpretation. He has also asked to start with (a) only.
 
  • #5
Orodruin said:
He has also asked to start with (a) only.
I missed that :rolleyes: .
 
  • #6
BvU said:
I missed that :rolleyes: .
It happens. I just pointed it out to highlight it.
 
  • #7
Orodruin said:
It happens. I just pointed it out to highlight it.
What happens ?
You obliging someone to start with a) ?
Or me missing that requirement in the original post ?
Even if you pointed it out there, I still don't see the necessity :mad: of starting with a).
I am hopeful @cookiemnstr510510 can do this on his/her own once the initial step is taken.
 
  • #8
Ah, misunderstanding. He/she wasn't asked, he/she has asked. My bad.

I feel free to point to b) to get started.
 
  • #9
Orodruin said:
i(d/dx) is a (linear) operator on your function space. What do you get of you act with it on a general function in your function space?

I believe I understand what you are saying, which is the inner product space is made of up all linear combinations of sin(x) and cos(x) which, like you said, is a function of the form f(x)=Asin(x)+Bcos(x). I am assuming A and B can be anything? complex or real? That makes sense to me, I went on Desmos and graphed about ten cosine and sine functions with different coefficients out front so I have a visual representation of what we mean.

Now to answer the quote above if we have an operator O=i(d/dx), this operator takes in a function, takes the partial derivative of it and multiplies it by i. So it might look something like this:
O(f(x))=O(sin(x))+O(cos(x))=i##\frac{d}{dx}(sin(x))##+i##\frac{d}{dx}(cos(x)##

or does our operator act on sin(x) and cos(x) separately, like this:
f(x)=sin(x), g(x)=cos(x)
O(f(x))=(i##\frac{d}{dx}(sin(x)))##+cos(x)
O(g(x))=(i##\frac{d}{dx}(cos(x)))##+sin(x)

Thanks for your help! It's a struggle learning what is being asked and how to approach.
 
  • #10
cookiemnstr510510 said:
I am assuming A and B can be anything? complex or real? That makes sense to me, I went on Desmos and graphed about ten cosine and sine functions with different coefficients out front so I have a visual representation of what we mean.
In general, you can make this into a real or complex vector space depending on what you intend. However, based on the rest of the problem, the intended vector space is over the complex numbers.

The operator acts on any function f(x) such that Of(x) = i f’(x)
 
  • Like
Likes cookiemnstr510510
  • #11
cookiemnstr510510 said:
or does our operator act on sin(x) and cos(x) separately, like this:
f(x)=sin(x), g(x)=cos(x)
O(f(x))=(i##\frac{d}{dx}(sin(x)))##+cos(x)
O(g(x))=(i##\frac{d}{dx}(cos(x)))##+sin(x)

Thanks for your help! It's a struggle learning what is being asked and how to approach.

Note that ##\frac{d}{dx}## is just the derivative. It's not any different just because we are now calling it an operator. If you take a function and differentiate it you get another function. That's what an operator does. So ##\frac{d}{dx}## has always been an operator, mapping differentiable functions into their derivatives.

If you want to know how ##O = i \frac{d}{dx}## acts on any function, just ask yourself how do I differentiate a function (and then multiply it by ##i##)?

In your example:

##i\frac{d}{dx} (\sin x) = i \cos x##

That's all there is to it.
 
  • Like
Likes cookiemnstr510510
  • #12
PeroK said:
Note that ##\frac{d}{dx}## is just the derivative. It's not any different just because we are now calling it an operator. If you take a function and differentiate it you get another function. That's what an operator does. So ##\frac{d}{dx}## has always been an operator, mapping differentiable functions into their derivatives.

If you want to know how ##O = i \frac{d}{dx}## acts on any function, just ask yourself how do I differentiate a function (and then multiply it be ##i##)?

In your example:

##i\frac{d}{dx} (\sin x) = i \cos x##

That's all there is to it.
Ahh, okay. This is starting to make more sense.

So since we are using sin(x) and cos(x) as our basis functions, and we are operating on them, we just differentiate each piece and multiply by i.

I understand that the inner product space defined is made from the linear combination of sin(x) and cos(x), which @Orodruin helped me understand is represented by f(x)=Asin(x)+Bcos(x). The problem then shows me the form of an inner product for functions (the bra ket = integral...).
So when the problem statement says "the inner product space consisting of all linear combinations of sin(x) and cos(x)" and gives me the "bra" "ket" integral, are these saying the same thing? Are we saying that the linear combination of sin(x) and cos(x) is representing the exact same thing as the bra-ket integral?

And let's say I wanted to solve this integral, would my |f>=sin(x) and |g>=cos(x)?

Thanks
 
  • #13
cookiemnstr510510 said:
Ahh, okay. This is starting to make more sense.

So since we are using sin(x) and cos(x) as our basis functions, and we are operating on them, we just differentiate each piece and multiply by i.

I understand that the inner product space defined is made from the linear combination of sin(x) and cos(x), which @Orodruin helped me understand is represented by f(x)=Asin(x)+Bcos(x). The problem then shows me the form of an inner product for functions (the bra ket = integral...).
So when the problem statement says "the inner product space consisting of all linear combinations of sin(x) and cos(x)" and gives me the "bra" "ket" integral, are these saying the same thing? Are we saying that the linear combination of sin(x) and cos(x) is representing the exact same thing as the bra-ket integral?

And let's say I wanted to solve this integral, would my |f>=sin(x) and |g>=cos(x)?

Thanks

Yes. Once you have established that functions have the properties of vectors (function spaces are an important example of vector spaces) and once you have defined an inner product on your function space:$$\langle f|g\rangle = \int f(x)^*g(x)dx$$ then you are free to use whatever notation you use to do your linear algebra, including the Dirac notation.

If you were studying pure maths then you would have to prove that this integral has the properties of an inner product. In QM, you might want to take time out to prove things like that or just get on with it!

One of the advantages of pure mathematics is that once you have proved that the integral behaves like an inner product, then you can do a lot of things generically without going back to an integral each time.

There is another major advantage of using the bra-ket notation that you'll come to soon. For now, yes it does represent the integral.

What you have here is:

1) A function space (with an inner product) consisting of all linear combinations of ##\cos x## and ##\sin x##. You could write this as$$V = \{a\sin x + b\cos x: \ a, b \in \mathbb{C}\}$$

2) A vector space (with an inner product) consisting of all vectors of the form $$V' = \{a|f\rangle + b|g \rangle: \ a, b \in \mathbb{C}\}$$ Where ##|f\rangle, g \rangle## represent ##\sin x, \cos x##.

What you are doing mathematically is setting up a correspondence between ##V## and ##V'## so you can switch back and forwards between them.
 
  • Like
Likes cookiemnstr510510
  • #14
cookiemnstr510510 said:
Are we saying that the linear combination of sin(x) and cos(x) is representing the exact same thing as the bra-ket integral?
No, the linear combination represents a vector in the function space. The braket integral represents the inner product between two vectors.
 
  • Like
Likes cookiemnstr510510
  • #15
With basis vectors ##\sin(x)## and ##\cos(x)##, any linear combination ##f(x) = A\sin(x)+B\cos(x)## is represented by the vector ##[A,B]^T##. Can you give the matrix that, operated on ##[A,B]^T## would give the derivative of ##f(x)## in that representation?
 
  • Like
Likes cookiemnstr510510
  • #16
FactChecker said:
With basis vectors ##\sin(x)## and ##\cos(x)##, any linear combination ##f(x) = A\sin(x)+B\cos(x)## is represented by the vector ##[A,B]^T##. Can you give the matrix that, operated on ##[A,B]^T## would give the derivative of ##f(x)## in that representation?

If I understand what you are saying correctly, I believe it would be:
[A,B]^T=[A;B]
[what matrix]⋅[A;B]=Acos(x)-Bsin(x)

[cos(x),0; 0 -sin(x)]⋅[A;B]
 
  • #17
cookiemnstr510510 said:
If I understand what you are saying correctly, I believe it would be:
[A,B]^T=[A;B]
[what matrix]⋅[A;B]=Acos(x)-Bsin(x)

[cos(x),0; 0 -sin(x)]⋅[A;B]
Close, but not exactly. ##d/dx [A\sin(x)+B\cos(x)] = A\cos(x)-B\sin(x) = -B\sin(x)+A\cos(x)## So there you want a matrix, ##M##, such that ##M[A;B] = [-B;A]##. You don't need to mention ##sin(x)## and ##cos(x)## in the definition of ##M##, since they are the basis vectors.
 

1. What is an inner product space?

An inner product space is a mathematical concept that refers to a vector space equipped with an inner product, which is a generalization of the dot product in Euclidean space. This inner product allows for the definition of angles and lengths of vectors, as well as the concept of orthogonality.

2. How is an inner product space represented using matrices?

An inner product space can be represented using matrices through the use of orthogonal bases and the Gram-Schmidt process. This process involves finding an orthonormal basis for the vector space and then constructing a matrix that represents the inner product using this basis.

3. What are the applications of inner product spaces?

Inner product spaces have a wide range of applications in mathematics, physics, and engineering. They are used to study vector spaces, linear transformations, and eigenvalues and eigenvectors. They also have practical applications in fields such as signal processing, quantum mechanics, and optimization problems.

4. What is the relationship between inner product spaces and operators?

An operator in an inner product space is a linear transformation from that space to itself. The inner product structure allows for the definition of adjoint operators, which are related to the original operator by a conjugate transpose. This relationship is important in applications such as quantum mechanics and signal processing.

5. How can inner product spaces be generalized?

Inner product spaces can be generalized in several ways, such as using non-Euclidean inner products or infinite-dimensional vector spaces. The concept of a Hilbert space, which is a complete inner product space, is also a generalization of the concept of an inner product space. These generalizations have applications in fields such as functional analysis and differential geometry.

Similar threads

  • Advanced Physics Homework Help
Replies
8
Views
369
  • Quantum Physics
Replies
8
Views
2K
  • General Math
Replies
7
Views
890
  • Advanced Physics Homework Help
Replies
1
Views
767
  • Calculus
Replies
8
Views
859
  • Calculus and Beyond Homework Help
Replies
5
Views
882
  • Advanced Physics Homework Help
Replies
3
Views
1K
  • Linear and Abstract Algebra
Replies
8
Views
976
  • Precalculus Mathematics Homework Help
Replies
5
Views
1K
  • Quantum Physics
Replies
2
Views
1K
Back
Top