Dot product, inner product, and projections

In summary, the dot product, also known as the inner product, is a mathematical operation that combines two vectors to produce a scalar, representing the extent to which the vectors align. It is computed by multiplying corresponding components of the vectors and summing the results. Projections involve using the dot product to determine how much one vector extends in the direction of another, allowing for geometric interpretations in vector spaces. This concept is essential in various applications, including physics, computer graphics, and machine learning, to analyze relationships between vectors.
  • #1
nomadreid
Gold Member
1,731
231
TL;DR Summary
Letting u,v be unit vectors, the length of the projection of u onto v is u dot v, whereas the inner product <u|v> is the projection of v onto u. Why the difference?
In simple Euclidean space: From trig, we have , for u and v separated by angle Θ, the length of the projection of u onto v is |u|cosΘ; then from one definition of the dot product Θ=arcos(|u|⋅|v|/(uv)); putting them together, I get the length of the projection of u onto v is uv/|v|.
Then I read that the inner product <u|v> is the result of the projection of v onto u.
Of course one could just say that the dot product is commutative, but the reverse order of what is projecting onto what seems a bit odd.
Either: where is my mistake, or: What am I missing?
Thanks in advance.
 
Mathematics news on Phys.org
  • #2
There are a ton of different sign and notation differences in math and physics. The best that you can hope for is that any given book or article is consistent. Even that is sometimes violated and a book/article notation convention may be dependent on the context.
 
  • Like
Likes nomadreid
  • #3
If you are interested in the topic you might want to read
https://arxiv.org/pdf/1205.5935.pdf

It is mathematics, but well written and a nice overview.
 
  • Like
Likes FactChecker
  • #4
nomadreid said:
TL;DR Summary: Letting u,v be unit vectors, the length of the projection of u onto v is u dot v, whereas the inner product <u|v> is the projection of v onto u.
That's a question of semantics. For me, ##\mathbf{u} \cdot \mathbf{v}## is projection of ##\mathbf{v}## on ##\mathbf{u}##, not the other way around. For real-valued vectors, there is no difference because of commutativity. For complex-valued vectors, it matters because the two inner products are complex conjugate of each other,
$$
\braket{u | v} = \overline{\braket{v | u}}
$$
Note tat another common notation for an inner product is ##(u,v)##, for which the convention is most often that ##v## is the quantity that will be complex-conjugated.
 
  • Like
Likes nomadreid
  • #5
Thanks, FactChecker, fresh_42 and DrClaude.
fresh_42: The book looks very clearly laid out, and I have downloaded it, as it will certainly be helpful.

DrClaude: I believe you have a typo in your note that the two inner products are complex conjugates of one another: there should be a line over one of the pair, or an asterisk, or however one chooses to indicate the complex conjugate.
 
  • #6
nomadreid said:
DrClaude: I believe you have a typo in your note that the two inner products are complex conjugates of one another: there should be a line over one of the pair, or an asterisk, or however one chooses to indicate the complex conjugate.
There is an overline. Maybe it is a question of MathJax rendering. What I see is
1696950897342.png
 
  • #7
DrClaude said:
There is an overline. Maybe it is a question of MathJax rendering. What I see is
View attachment 333411
I see the same, both here and in your previous post.
 
  • #8
Mark44 said:
I see the same, both here and in your previous post.
On my Windows 10 PC Firefox browser, I don't see that in the post, only in the .png image.
On the Chrome browser, I see it correctly in the post.
On my Samsung Android tablet Chrome browser, I see it correctly in the post.
 

FAQ: Dot product, inner product, and projections

What is the dot product?

The dot product, also known as the scalar product, is an algebraic operation that takes two equal-length sequences of numbers (usually coordinate vectors) and returns a single number. It is calculated as the sum of the products of their corresponding components. For example, for two vectors A and B in n-dimensional space, the dot product is given by A · B = A1 * B1 + A2 * B2 + ... + An * Bn.

How is the dot product related to the angle between two vectors?

The dot product can be used to find the cosine of the angle θ between two vectors A and B. The relationship is given by the formula A · B = |A| |B| cos(θ), where |A| and |B| are the magnitudes (lengths) of the vectors. By rearranging this equation, you can find the angle: θ = cos⁻¹((A · B) / (|A| |B|)).

What is the inner product?

The inner product is a generalization of the dot product that can be defined in any inner product space. It satisfies certain properties, including linearity in the first argument, symmetry (or conjugate symmetry in complex spaces), and positive definiteness. The inner product provides a way to define angles and lengths in more abstract vector spaces.

How do you calculate the projection of one vector onto another?

The projection of a vector A onto another vector B is calculated using the formula: proj_B(A) = (A · B / B · B) * B. This formula gives you a vector that represents the orthogonal projection of A onto B, indicating how much of A lies in the direction of B.

What are the applications of dot products and projections in real life?

Dot products and projections have numerous applications in various fields. In physics, they are used to calculate work done by a force. In computer graphics, they help determine lighting and shading by calculating angles between light sources and surfaces. In machine learning, dot products are used in algorithms for classification and clustering, while projections are used in dimensionality reduction techniques like Principal Component Analysis (PCA).

Similar threads

Replies
4
Views
2K
Replies
4
Views
2K
Replies
4
Views
2K
Replies
2
Views
2K
Replies
8
Views
3K
Replies
4
Views
2K
Replies
4
Views
1K
Back
Top