ZhouYuxuanYX/Maximum-Suppression-Regularization
This is the official repository of our NeurIPS 2025 paper "MaxSup: Overcoming Representation Collapse in Label Smoothing"
This project offers a new regularization technique, MaxSup, to enhance how classification models learn from data. It takes labeled image datasets and outputs a more robustly trained model capable of better classification and feature extraction. Data scientists, machine learning engineers, and AI researchers working on computer vision tasks would use this to improve model performance.
Use this if you are training image classification models and want to improve their performance, prevent overconfident errors, and achieve better transfer learning results compared to traditional label smoothing.
Not ideal if your primary goal is not image classification or if you are not working with deep learning models that benefit from regularization techniques.
Stars
22
Forks
4
Language
Jupyter Notebook
License
—
Category
Last pushed
Nov 06, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ZhouYuxuanYX/Maximum-Suppression-Regularization"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
AdaptiveMotorControlLab/CEBRA
Learnable latent embeddings for joint behavioral and neural analysis - Official implementation of CEBRA
theolepage/sslsv
Toolkit for training and evaluating Self-Supervised Learning (SSL) frameworks for Speaker...
PaddlePaddle/PASSL
PASSL包含 SimCLR，MoCo v1/v2，BYOL，CLIP，PixPro，simsiam, SwAV, BEiT，MAE 等图像自监督算法以及 Vision...
YGZWQZD/LAMDA-SSL
30 Semi-Supervised Learning Algorithms
ModSSC/ModSSC
ModSSC: A Modular Framework for Semi Supervised Classification