add method: Uniform Risk Minimization (URM)#15
Open
kiranchari wants to merge 2 commits intoYyzHarry:mainfrom
Open
add method: Uniform Risk Minimization (URM)#15kiranchari wants to merge 2 commits intoYyzHarry:mainfrom
kiranchari wants to merge 2 commits intoYyzHarry:mainfrom
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
I am first author of "Uniformly Distributed Feature Representations for Fair and Robust Learning" (TMLR 2024) paper. Link: https://openreview.net/forum?id=PgLbS5yp8n
In this pull request I am adding code for the proposed Uniform Risk Minimization (URM) method for subpopulation robustness. The key idea is to encourage the distribution of feature representations learned by the encoder/featurizer to be uniformly distributed. The paper provides theoretical and empirical support for the proposed method in sub-population robustness and also domain generalization.
Changes include:
URMclass inalgorithms.pyhparams_registry.pynetworks.pyandlib/wide_resnet.pyto add an activation function to the output of the featurizer/encoder networks (before the linear classifier). This does not affect other algorithms in SubpopBench as the default activation is nn.Identity(): https://github.com/kiranchari/SubpopBench/blob/198ea9548948b5af0a084798878c880e89d248d6/subpopbench/models/networks.py#L31README.mdThank you.