Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Second derivative and tensors

  1. Nov 26, 2013 #1
    Given a function f(x(t, s) y(t, s)), if is possible to compact
    [tex]\frac{∂f}{∂t}=\frac{∂f}{∂x} \frac{∂x}{∂t}+\frac{∂f}{∂y} \frac{∂y}{∂t}[/tex]
    [tex]\frac{df}{dt}=\bigtriangledown f\cdot \frac{d\vec{r}}{dt}[/tex]

    So, analogously, isn't possible to compact the sencond derivate
    [tex]\frac{\partial^2 f}{\partial s \partial t} = \frac{\partial^2 f}{\partial x^2} \frac{\partial x}{\partial s }\frac{\partial x}{\partial t } + \frac{\partial^2 f}{\partial x \partial y}\left( \frac{\partial y}{\partial s }\frac{\partial x}{\partial t } + \frac{\partial x}{\partial s }\frac{\partial y}{\partial t }\right) + \frac{\partial^2 f}{\partial y^2} \frac{\partial y}{\partial s }\frac{\partial y}{\partial t }+\frac{\partial f}{\partial x}\frac{\partial^2 x}{\partial s \partial t}+\frac{\partial f}{\partial y}\frac{\partial^2 y}{\partial s \partial t}[/tex]
    using matrix with the matrix Hessian(f) ?
  2. jcsd
  3. Nov 27, 2013 #2
    [tex]\frac{\partial}{\partial t} = \sum_{i} \frac{\partial x_i}{\partial t}\frac{\partial}{\partial x_i} [/tex]
    [tex]\frac{\partial}{\partial s} = \sum_{i} \frac{\partial x_i}{\partial s}\frac{\partial}{\partial x_i} [/tex]
    [tex]\frac{\partial^2}{\partial s\partial t}f = \sum_{i,j} \frac{\partial x_i}{\partial s}\frac{\partial}{\partial x_i} \frac{\partial x_j}{\partial t}\frac{\partial f}{\partial x_j} [/tex]
    Next I applied the product rule:
    [tex]\frac{\partial^2}{\partial s\partial t}f = \sum_{i,j} \frac{\partial x_i}{\partial s}\left( \frac{\partial^2 x_j}{\partial x_i \partial t}\frac{\partial f}{\partial x_j} +\frac{\partial x_j}{\partial t}\frac{\partial f}{\partial x_i \partial x_j} \right) [/tex]
    The second term above involves the Hessian matrix, which is sandwiched between [itex]\partial \mathbf{x}^T/\partial s[/itex] and [itex]\partial \mathbf{x}/\partial t[/itex]. I've moved this to become the first term.
    [tex]\frac{\partial^2}{\partial s\partial t}f = \sum_{i,j} \frac{\partial x_i}{\partial s} H_{ij}(f)\frac{\partial x_j}{\partial t} + \sum_{i,j}\frac{\partial x_i}{\partial s}\frac{\partial^2 x_j}{\partial x_i \partial t}\frac{\partial f}{\partial x_j} [/tex]
    Finally I reverse chain ruled the second term.
    [tex]\frac{\partial^2}{\partial s\partial t}f = \sum_{i,j} \frac{\partial x_i}{\partial s} H_{ij}(f)\frac{\partial x_j}{\partial t} + \sum_{j}\frac{\partial^2 x_j}{\partial s \partial t}\frac{\partial f}{\partial x_j} [/tex]

    You may want to check my math.
  4. Nov 27, 2013 #3
    Your math appears to be correct. But, I don't know how you and outhers are capable to comprehend this summation notation!

    I'm looking for matrix, without summation.
  5. Nov 27, 2013 #4
    How's this?

    [tex]\frac{\partial^2 f}{\partial s \partial t} = \frac{\partial \mathbf{x}^T}{\partial s} \mathbf{H}(f)\frac{\partial \mathbf{x}}{\partial t} + \frac{\partial^2 \mathbf{x}}{\partial s \partial t} \cdot \nabla f[/tex]
  6. Nov 27, 2013 #5
    Oh yeah! This is what I was looking for! I never would be capable to deduce this!
    By the way, it's complicated!
  7. Dec 5, 2013 #6
    Moreover... in matricial form, could be wrote so:
    [tex]\begin{bmatrix} \frac{\partial^2 f}{\partial x\partial x} & \frac{\partial^2 f}{\partial x\partial y}\\ \frac{\partial^2 f}{\partial y\partial x} & \frac{\partial^2 f}{\partial y\partial y} \end{bmatrix} \cdot \begin{bmatrix} \frac{\partial x}{\partial s} \frac{\partial x}{\partial t} & \frac{\partial x}{\partial s} \frac{\partial y}{\partial t}\\ \frac{\partial y}{\partial s} \frac{\partial x}{\partial t} & \frac{\partial y}{\partial s} \frac{\partial y}{\partial t} \end{bmatrix} + \begin{bmatrix} \frac{\partial f}{\partial x} \\ \frac{\partial f}{\partial y} \end{bmatrix} \cdot \begin{bmatrix} \frac{\partial^2 x}{\partial s\partial t}\\ \frac{\partial^2 y}{\partial s\partial t} \end{bmatrix}[/tex]

    But, I ask... exist dot product between matrices (tensors of rank 2)?

    And is possible call the second matrix of something more simplificated? The first is H(f), the third is ∇f and the fourth is ∂²x/∂s∂t...
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook