Function taking vectors of Rn as inputs

  • Context: Graduate 
  • Thread starter Thread starter gursimran
  • Start date Start date
  • Tags Tags
    Function Vectors
Click For Summary
SUMMARY

A function can take inputs as vectors in Rn and return a real value R, exemplified by the dot product. This function can be defined as f(Rn, Rn) -> R, which is a bilinear function. To solve the equation |c - x·y| using gradient descent, where c is a scalar and x, y are vectors in Rn, one can analyze the expression |∑(ck - xkyk)|. The graph of this function exists in R2n+1, limiting visualizations to n = 1, which may not provide comprehensive insights but illustrates the principle.

PREREQUISITES
  • Understanding of vector spaces in Rn
  • Familiarity with bilinear functions
  • Knowledge of gradient descent optimization techniques
  • Basic graph visualization in higher dimensions
NEXT STEPS
  • Study the properties of bilinear functions in Rn
  • Learn about gradient descent algorithms and their applications
  • Explore visualization techniques for functions in R2n and R2n+1
  • Investigate the mathematical foundations of Singular Value Decomposition (SVD)
USEFUL FOR

Mathematicians, data scientists, and machine learning practitioners interested in optimization techniques and function visualization in multi-dimensional spaces.

gursimran
Messages
35
Reaction score
0
Hello friends,

Is it possible that a function take inputs as vectors in Rn and return one real value R. One such example is dot product. But keeping mathematical rigor in mind what should I call this function.

Problem 1.
Its a function from R2n to R
or
a function such that f(Rn,Rn)->R

Problem 2.
I want to solve this equation using gradient descent
|c - x.y|

here
c - scalar (R)
x and y are vectors in Rn

How should I visualize its graph?

This problem is inspired by the mathematical treatment used to calculated SVD here.

As proved by Ecart and young there could only be one minimum of the equation and the minimum should be SVD. So it means it should be a sort of parabolloid. How to visualize it.
 
Physics news on Phys.org
gursimran said:
Hello friends,

Is it possible that a function take inputs as vectors in Rn and return one real value R. One such example is dot product. But keeping mathematical rigor in mind what should I call this function.

Problem 1.
Its a function from R2n to R
or
a function such that f(Rn,Rn)->R
If you want to consider it a bilinear function as usual, then it is the latter. If you are only interested in analytic properties, then you can use the former, but then we no longer have linearity.
Problem 2.
I want to solve this equation using gradient descent
|c - x.y|

here
c - scalar (R)
x and y are vectors in Rn
You can analytically consider ##\left| \sum_k (c_k - x_ky_k) \right|= n \circ \sum_k (c_k - x_ky_k)## and use the definition of your norm ##n##.
How should I visualize its graph?
The graph of a function ##f\, : \,\mathbb{R}^{2n} \longrightarrow \mathbb{R}## is an object in ##\mathbb{R}^{2n+1}##. This constraints visualizations to ##n = 1##. Probably not very informative but at least it displays the principle.
This problem is inspired by the mathematical treatment used to calculated SVD here.

As proved by Ecart and young there could only be one minimum of the equation and the minimum should be SVD. So it means it should be a sort of parabolloid. How to visualize it.
 

Similar threads

  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 17 ·
Replies
17
Views
3K
  • · Replies 43 ·
2
Replies
43
Views
8K
  • · Replies 10 ·
Replies
10
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 25 ·
Replies
25
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 17 ·
Replies
17
Views
4K
  • · Replies 8 ·
Replies
8
Views
2K