Sketch the gradient vector for the function

Click For Summary

Homework Help Overview

The discussion revolves around sketching the gradient vector for a function, focusing on its direction and magnitude in relation to level curves.

Discussion Character

  • Exploratory, Conceptual clarification, Mathematical reasoning

Approaches and Questions Raised

  • Participants explore the direction of the gradient as perpendicular to the tangent of the level curve and question how to determine its length. There is discussion about approximating the gradient's magnitude based on the distance between level curves and the increase in function values.

Discussion Status

Participants are actively engaging with the concepts, attempting to clarify their understanding of the gradient's properties. Some guidance is offered regarding the relationship between rise and run in the context of gradient estimation, but there is no explicit consensus on the correct approach.

Contextual Notes

There is mention of specific points on the graph and the need to estimate distances between level curves, which may vary depending on the location in question. Participants express confusion about the concept of gradient and its application.

Feodalherren
Messages
604
Reaction score
6

Homework Statement


Untitled.png



Homework Equations





The Attempt at a Solution


Ok so I think I know how to get the direction. It's going to be perpendicular to the tangent of the level curve and pointing in the direction where f(x,y) is increasing. So on the graph that was provided it will point in negative y and positive x.

I am completely lost on finding the length though.
 
Physics news on Phys.org
Feodalherren said:

Homework Statement


Untitled.png



Homework Equations





The Attempt at a Solution


Ok so I think I know how to get the direction. It's going to be perpendicular to the tangent of the level curve and pointing in the direction where f(x,y) is increasing. So on the graph that was provided it will point in negative y and positive x.

I am completely lost on finding the length though.

The length of the gradient is going to proportional to the rate at which f(x,y) is increasing as you move along the direction of the gradient. Can't you approximate that by the distance between the level curves and the amount f(x,y) increases between them?
 
The distance between level curves -2 and -1 = 1

Isn't that how much f(x,y) increases?
 
Feodalherren said:
The distance between level curves -2 and -1 = 1

Isn't that how much f(x,y) increases?

Right. So the magnitude of the gradient is going be approximately 1/(distance between the level curves in the direction of the gradient). The denominator will depend on where you are on the level curves.
 
Ok so in a simple rise/run, that means 1 is my rise, what is my run?

My professor has the right answer as "approx 2".
 
Feodalherren said:
Ok so in a simple rise/run, that means 1 is my rise, what is my run?

My professor has the right answer as "approx 2".

At (4,6)? The run is the perpendicular distance between the level curves at that point. What's that "approx"?
 
I guess it stands for "approximately".

This just isn't working out at all.. I'm so confused :/
 
Feodalherren said:
I guess it stands for "approximately".

This just isn't working out at all.. I'm so confused :/

I guessed what "approx" meant, I was just asking you what your guess for the run was around the point (4,6). What's your guess for the distance between the level curves -1 and -2 near that point by reading off the graph? It is just a rise/run guesstimate.
 
Well my guess is that the rise is 1.

As for the run that's where I get lost. Do i estimate in the same direction as the gradient, or in the X or Y direction? What's the deal here?
 
  • #10
Feodalherren said:
Well my guess is that the rise is 1.

As for the run that's where I get lost. Do i estimate in the same direction as the gradient, or in the X or Y direction? What's the deal here?

In the same direction as the gradient, i.e. perpedicular to the level curves. Looks to me like about 1/2 is a good guess around (4,6). Don't you agree?
 
  • Like
Likes   Reactions: 1 person
  • #11
Ah it makes sense in that case. I would have guessed that it would be in that direction too but I wasn't sure. I guess I'm just not really grasping the concept. I realize that a gradient is just essentially a derivative in any direction but for some reason I'm having a hard time wrapping my head around this.
Anyway, thanks a bunch for your patience!
 

Similar threads

  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
Replies
12
Views
9K
Replies
3
Views
2K
  • · Replies 10 ·
Replies
10
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K