-
Notifications
You must be signed in to change notification settings - Fork 17
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Darcy example documentation #318
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I left some comments, please let me know what you think!
@@ -0,0 +1,179 @@ | |||
# # [Learning the Pearmibility field in a Darcy flow from noisy sparse observations] | |||
|
|||
# In this example we hope to illustrate function learning. One may wish to use function learning in cases where the underlying parameter of interest is actual a finite-dimensional approximation (e.g. spatial discretization) of some "true" function. Treating such an object directly will lead to increasingly high-dimensional learning problems as the spatial resolution is increased, resulting in poor computational scaling and increasingly ill-posed inverse problems. Treating the object as a discretized function from a function space, one can learn coefficients not in the standard basis, but instead in a basis of this function space, it is commonly the case that functions will have relatively low effective dimension, and will be depend only on the spatial discretization due to discretization error, that should vanish as resolution is increased. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
small typo: "parameter of interest is actual a" -> "parameter of interest is actually a".
My interpretation of this paragraph is as follows:
In this example, we hope to illustrate function learning. While traditional inverse problems involve the calibration of parameters of a predefined model, function learning allows one to learn a finite-dimensional approximation (e.g. spatial discretization) of some “true” function/model. When treating the approximation as just random samples at evaluation points, increasing the spatial resolution leads to increasingly high dimensional learning problems, thus giving poor computational scaling and increasingly ill-posed inverse problems. When treating the approximation instead as a discretized function from a function space, one can learn coefficients instead of a basis in this function space. Since it is commonly the case that functions have relatively low effective dimension in this space, the dependence on the spatial discretization only arises in discretization error, which vanishes as resolution is increased.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I did not quite insert this, interpretation, but I hope that the new version is a good intermediary. Feel free to iterate again
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Also note that this comment was left on darcy.jl example, not on the darcy.md docs
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I like the wording here, it is much clearer what the example is doing.
bors r+ |
Build succeeded! The publicly hosted instance of bors-ng is deprecated and will go away soon. If you want to self-host your own instance, instructions are here. If you want to switch to GitHub's built-in merge queue, visit their help page. |
Purpose
Resolves #317
Content
example/Darcy
See the docs page here