Name: Curtis G. Northcutt
Type: User
Company: @Cleanlab | MIT
Bio: CEO & Co-Founder @cleanlab | PhD @MIT | Creator of Cleanlab package | Former Google, Oculus, Amazon, Facebook, Microsoft, NASA AI
Twitter: cgnorthcutt
Location: Cambridge, MA
Blog: https://www.curtisnorthcutt.com/
Curtis G. Northcutt's Projects
A curated list of resources for Learning with Noisy Labels
🔥 Reproducibly benchmarking Keras and PyTorch models
A Python Pandas library for cheating, collaboration, proficiency, and retention detection in edX courses ingested with Google BigQuery.
Curtis G. Northcutt's personal website.
Official cleanlab repo is at https://github.com/cleanlab/cleanlab
Latest (2020) CNN and GPU Benchmarks on ImageNet and CIFAR
NeurIPS'18: Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labels
Official data release to reproduce Confident Learning paper results
ICML'19: How does Disagreement Help Generalization against Label Corruption?
cuML - RAPIDS Machine Learning Library
A Jekyll-based theme designed for documentation and help systems. See the link for detailed instructions on setting up and configuring everything.
Uses dotbot to handle my dotfiles.
EgoCom: A Multi-person Multi-modal Egocentric Communications Dataset
A set of examples around pytorch in Vision, Text, Reinforcement Learning, etc.
Add diversity to the order of comments in forums!
A collection of useful .gitignore templates
"Best Jekyll Theme by a Mile"
⏸ Parallelized hyper-param optimization with validation set, not crossval
IEEE Computer Society Keywords to Organize Knowledge
an imagenet example in torch.
Simpler human-readable labels for ImageNet 🏷
Sum-safe Rounding for Iterables
Markdown to reStructuredText converter
mixup: Beyond Empirical Risk Minimization
Models and examples built with TensorFlow
ICML 2019: Understanding and Utilizing Deep Neural Networks Trained with Noisy Labels
The official website of PomDP the PhD rapper.
Fork this repo for a quick start. If "Project Timeline" or "License" appeared on your nav bar, Look Below!