RylanSchaeffer/Stanford-AI-Alignment-Double-Descent-Tutorial
Code for Arxiv Double Descent Demystified: Identifying, Interpreting & Ablating the Sources of a Deep Learning Puzzle
This project helps machine learning researchers and practitioners understand the phenomenon of "double descent" in deep learning models. By analyzing various regression models, it provides insights into why model performance can initially worsen and then improve again with increasing complexity. It takes raw data and model configurations as input and outputs visualizations and analyses that clarify the sources of this puzzling behavior. This is ideal for those actively researching or implementing advanced machine learning models.
No commits in the last 6 months.
Use this if you are a machine learning researcher or advanced practitioner investigating model generalization, overfitting, and the 'double descent' phenomenon in deep learning.
Not ideal if you are a beginner looking for a simple machine learning tutorial or if your primary goal is to build and deploy production-ready models without deep theoretical investigation.
Stars
66
Forks
10
Language
Python
License
—
Category
Last pushed
Nov 24, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/RylanSchaeffer/Stanford-AI-Alignment-Double-Descent-Tutorial"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
pykt-team/pykt-toolkit
pyKT: A Python Library to Benchmark Deep Learning based Knowledge Tracing Models
microsoft/archai
Accelerate your Neural Architecture Search (NAS) through fast, reproducible and modular research.
google-research/morph-net
Fast & Simple Resource-Constrained Learning of Deep Network Structure
AI-team-UoA/pyJedAI
An open-source library that leverages Python’s data science ecosystem to build powerful...
IDEALLab/EngiBench
Benchmarks for automated engineering design