yangarbiter/dp-dg
What You See is What You Get: Distributional Generalization for Algorithm Design in Deep Learning
This project provides methods for designing deep learning models that perform predictably on new, unseen data, based on their training performance. It takes standard deep learning models and training datasets, and outputs models with more reliable real-world generalization. Data scientists and machine learning engineers can use this to build more robust and fair AI systems.
No commits in the last 6 months.
Use this if you want to ensure your deep learning model's behavior on training data closely matches its behavior on new, real-world data, especially for applications sensitive to fairness or robustness.
Not ideal if your primary concern is solely maximizing predictive accuracy without regard for distributional generalization or fairness implications.
Stars
7
Forks
1
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Oct 05, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/yangarbiter/dp-dg"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
google/scaaml
SCAAML: Side Channel Attacks Assisted with Machine Learning
pralab/secml
A Python library for Secure and Explainable Machine Learning
Koukyosyumei/AIJack
Security and Privacy Risk Simulator for Machine Learning (arXiv:2312.17667)
AI-SDC/SACRO-ML
Collection of tools and resources for managing the statistical disclosure control of trained...
oss-slu/mithridatium
Mithridatium is a research-driven project aimed at detecting backdoors and data poisoning in...