loss-correction
Loss correction framework
Provides a framework for implementing robust loss functions to mitigate the effects of label noise in deep neural networks.
Robust loss functions for deep neural networks (CVPR 2017)
88 stars
8 watching
18 forks
Language: Python
last commit: over 4 years ago Related projects:
| Repository | Description | Stars |
|---|---|---|
 hanxunh/active-passive-losses | A PyTorch-based framework for implementing normalized loss functions to improve deep learning model robustness against noisy labels. | 134 |
 bermanmaxim/jaccardsegment | A deep learning framework implementing Deeplab-resnet-101 with binary Jaccard loss surrogate, based on the Lovász hinge loss. | 97 |
 marcoancona/deepexplain | A framework for understanding how deep neural networks process input data to produce output | 734 |
 mc-e/deep-generalized-unfolding-networks-for-image-restoration | An image restoration framework using neural networks with interpretable and adaptive structure for diverse applications | 131 |
 alanchou/truncated-loss | An implementation of a loss function designed to improve the training of deep neural networks with noisy labels | 125 |
 guopengf/auto-fedrl | A reinforcement learning-based framework for optimizing hyperparameters in distributed machine learning environments. | 15 |
 coincheung/pytorch-loss | Provides a comprehensive set of implementation of various loss functions and operators for deep learning models | 2,181 |
 paulalbert31/labelnoisecorrection | An implementation of an unsupervised label noise modeling and loss correction approach for deep learning. | 220 |
 tobypde/frrn | A software framework for training and evaluating full-resolution residual networks for semantic image segmentation tasks | 280 |
 xiaoyu258/geoproj | A deep learning-based approach to correcting geometric distortions in images | 187 |
 jonfanlab/glonet | A software framework for training neural networks to optimize dielectric metasurfaces using physics-driven generative models and global optimization algorithms. | 101 |
 ganyuwang/vfl-czofo | A unified framework for improving privacy and reducing communication overhead in distributed machine learning models. | 11 |
 kentonishi/augmentation-for-lnl | Provides a framework for learning with noisy labels using data augmentation strategies. | 113 |
 sarababakn/mfcl-neurips23 | A framework for mitigating catastrophic forgetting in federated learning for vision tasks using data synthesis from past distributions. | 15 |
 mmazeika/glc | A method to train deep learning classifiers on noisy labels using a small set of trusted data | 86 |