Implementations of different loss-correction techniques to help deep models learn under class-conditional label noise.
-
Updated
Feb 15, 2021 - Python
Implementations of different loss-correction techniques to help deep models learn under class-conditional label noise.
deep-learning image classification resnet50
Label Smoothing and Adversarial Robustness
Anime Face Generation using GANS and Label Smoothing.
PyTorch code for Sparse Label Smoothing Regularization presented in "Learning Symbolic Model-Agnostic Loss Functions via Meta-Learning" (TPAMI-2023). Paper Link: https://arxiv.org/abs/2209.08907
Code for "Memorization-Dilation: Modeling Neural Collapse under Noise" as published at ICLR 2023.
Multiple Generation Based Knowledge Distillation: A Roadmap
Modern Eager TensorFlow implementation of Attention Is All You Need
A simple template for classifying things
Adding Image-context in the Label Smoothing process via Geodesic distance
Building High Performance Convolutional Neural Networks with TensorFlow
Soft Target and Label Smoothing in Text Classification for Probability Calibration of Output Distributions.
📦Simple Tool Box with Pytorch
Build an algorithm that can predict multiple future states of Limit Order Books using high-frequency, multi-variate, short time-frame data
[ICML 2022] This work investigates the compatibility between label smoothing (LS) and knowledge distillation (KD). We suggest to use an LS-trained teacher with a low-temperature transfer to render high performance students.
Label Smoothing applied in Focal Loss
Mean Teacher-based Cross-Domain Activity Recognition using WiFi Signals, IoTJ 2023
Label smoothed Aggregation cross entropy loss for generalisation in sequence to sequence tasks.
Supplementary material and code for "From Label Smoothing to Label Relaxation" as published at AAAI 2021.
Source code of our paper "Focus on the Target’s Vocabulary: Masked Label Smoothing for Machine Translation" @acl-2022
Add a description, image, and links to the label-smoothing topic page so that developers can more easily learn about it.
To associate your repository with the label-smoothing topic, visit your repo's landing page and select "manage topics."