Using self-supervised learning can improve model robustness and uncertainty
D Hendrycks, M Mazeika… - Advances in neural …, 2019 - proceedings.neurips.cc
… Self-supervision provides effective … fully supervised training and are often not thought
beneficial beyond obviating or reducing the need for annotations. We find that self-supervision can …
beneficial beyond obviating or reducing the need for annotations. We find that self-supervision can …
Semantic segmentation with generative models: Semi-supervised learning and strong out-of-domain generalization
… In the semi-supervised setting, we demonstrate results equal … (iii) We show strong
generalization capabilities and outper… models: Weight-averaged consistency targets improve semi-…
generalization capabilities and outper… models: Weight-averaged consistency targets improve semi-…
Sharpness-aware minimization for efficiently improving generalization
… of the loss landscape and generalization, we … improves model generalization across a
variety of benchmark datasets (eg, CIFAR-10, CIFAR-100, ImageNet, finetuning tasks) and models…
variety of benchmark datasets (eg, CIFAR-10, CIFAR-100, ImageNet, finetuning tasks) and models…
Supervised contrastive learning
… Our loss can be seen as a generalization of both the triplet [… improve upon [41] by normalizing
the embeddings and replacing euclidean distance with inner products. We further improve …
the embeddings and replacing euclidean distance with inner products. We further improve …
Virtual adversarial training: a regularization method for supervised and semi-supervised learning
… was able to improve the generalization performance. Random … to improve generalization
performance for semi-supervised … of techniques that smooth the model around input data points …
performance for semi-supervised … of techniques that smooth the model around input data points …
Big self-supervised models are strong semi-supervised learners
… ) neural network for self-supervised pretraining and fine-tuning greatly improves accuracy. In
… , which could easily overfit with few labeled examples, can generalize much better. With task…
… , which could easily overfit with few labeled examples, can generalize much better. With task…
Unified deep supervised domain adaptation and generalization
S Motiian, M Piccirilli, DA Adjeroh… - Proceedings of the …, 2017 - openaccess.thecvf.com
… the base model, emphasizing the need for domain generalization. Our … model to provide
better performance. Figure 4(c) shows the improvement of our DG model over the base model …
better performance. Figure 4(c) shows the improvement of our DG model over the base model …
Improving generalization with active learning
… Since we are interested in selecting examples that improve the generalization behavior of
some given neural network architecture N, we define the concept class in question to be the …
some given neural network architecture N, we define the concept class in question to be the …
Semi-supervised support vector machines
… We show how the S3VM model for I-norm linear support vector … data improves generalization
when insufficient training information is available. In every case, S3VM either improved or …
when insufficient training information is available. In every case, S3VM either improved or …
Learning to generalize: Meta-learning for domain generalization
… Thus DG is to achieve an agent with improved generalization ability in the sense of … ) –
without being allowed any rewarded (≈ supervised domain adaptation (Finn, Abbeel, and Levine …
without being allowed any rewarded (≈ supervised domain adaptation (Finn, Abbeel, and Levine …
関連キーワード
- generative models semi-supervised learning
- improving the generalization of unsupervised models
- improving the generalization of machine learning
- improve the generalization of supervised models
- improving the generalization of neural networks
- improving the generalization of nlp models
- improving the generalizability of supervised models
- enhancing the generalization of supervised models
- improving the accuracy of supervised models
- improving the generalization of deep learning
- improving the generalization of regression models
- improving the generalization of transformer models
- improving the generalization of cnn models