Skip to content
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

create content for custom losses #35

Open
lisa-wm opened this issue Mar 2, 2021 · 0 comments
Open

create content for custom losses #35

lisa-wm opened this issue Mar 2, 2021 · 0 comments

Comments

@lisa-wm
Copy link
Contributor

lisa-wm commented Mar 2, 2021

currently only below teaser exists, which is not in any active slides.
some slides on defining a custom loss should be drawn up (with enough mathematical rigour).


Assume a use case, where the target variable can have a wide range of values across different orders of magnitude.
A possible solution would be to use a loss functions that allows for better model evaluation and comparison.
The \textbf{Mean Squared Logarithmic Absolute Error} is not strongly influenced by large values due to the logarithm.

[
\frac{1}{n} \sumin (\log(|\yi - \yih| + 1))^2
]

@mb706 mb706 transferred this issue from slds-lmu/lecture_i2ml Jan 28, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants