neurodata/value-of-ood-data
The value of out-of-distribution data (ICML 2023)
This project explores how adding data that is 'out-of-distribution' (OOD) impacts the accuracy of machine learning models. It shows that adding small amounts of OOD data can sometimes improve accuracy, but adding too much can actually make the model perform worse than if no OOD data was used at all. This research is for machine learning practitioners, data scientists, and researchers who are building and evaluating models with diverse datasets.
No commits in the last 6 months.
Use this if you are a machine learning researcher or data scientist investigating the optimal use of varied datasets, especially when some data might not perfectly match your main task.
Not ideal if you are looking for a plug-and-play solution to automatically filter or weigh OOD data in a production system.
Stars
10
Forks
—
Language
Python
License
—
Category
Last pushed
Oct 31, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/neurodata/value-of-ood-data"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
Trusted-AI/adversarial-robustness-toolbox
Adversarial Robustness Toolbox (ART) - Python Library for Machine Learning Security - Evasion,...
bethgelab/foolbox
A Python toolbox to create adversarial examples that fool neural networks in PyTorch, TensorFlow, and JAX
cleverhans-lab/cleverhans
An adversarial example library for constructing attacks, building defenses, and benchmarking both
DSE-MSU/DeepRobust
A pytorch adversarial library for attack and defense methods on images and graphs
BorealisAI/advertorch
A Toolbox for Adversarial Robustness Research