yfzhang114/Environment-Label-Smoothing

This is an official PyTorch implementation of the ICLR 2023 paper 《Free Lunch for Domain Adversarial Training: Environment Label Smoothing》.

23
/ 100
Experimental

This project helps machine learning engineers improve the accuracy and reliability of their models when applied to new, unseen datasets. It takes existing domain adversarial training models and applies a "label smoothing" trick to produce models that generalize better across different data environments. Data scientists and machine learning researchers dealing with real-world data variability will find this useful.

No commits in the last 6 months.

Use this if your machine learning models struggle to perform well when deployed on data that is slightly different from your training data, causing issues like inconsistent predictions or poor generalization.

Not ideal if you are not already using or planning to use domain adversarial training methods, as this method specifically enhances that approach.

machine-learning-generalization domain-adaptation image-classification natural-language-processing genomics
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 8 / 25
Community 7 / 25

How are scores calculated?

Stars

60

Forks

3

Language

Python

License

Last pushed

Feb 04, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/yfzhang114/Environment-Label-Smoothing"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.