Learning to Impute: A General Framework for Semi-supervised Learning
arXiv:1912.10364
Abstract
Recent semi-supervised learning methods have shown to achieve comparable results to their supervised counterparts while using only a small portion of labels in image classification tasks thanks to their regularization strategies. In this paper, we take a more direct approach for semi-supervised learning and propose learning to impute the labels of unlabeled samples such that a network achieves better generalization when it is trained on these labels. We pose the problem in a learning-to-learn formulation which can easily be incorporated to the state-of-the-art semi-supervised techniques and boost their performance especially when the labels are limited. We demonstrate that our method is applicable to both classification and regression problems including image classification and facial landmark detection tasks.
Semi-supervised Learning, Meta-Learning, Learning to impute
References in corpus (24)
- Adam: A Method for Stochastic Optimization
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- mixup: Beyond Empirical Risk Minimization
- Improved Regularization of Convolutional Neural Networks with Cutout
- FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence
- Unsupervised Data Augmentation for Consistency Training
- Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
- Learning to Reweight Examples for Robust Deep Learning
- Meta-Learning for Semi-Supervised Few-Shot Classification
- MixMatch: A Holistic Approach to Semi-Supervised Learning
- Learning multiple visual domains with residual adapters
- Learning Deep Representation for Face Alignment with Auxiliary Attributes
- QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehension
- Learning to Self-Train for Semi-Supervised Few-Shot Classification
- Meta-Learning with Implicit Gradients
- Adversarial Training Methods for Semi-Supervised Text Classification
- Unsupervised Learning via Meta-Learning
- ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring
- There Are Many Consistent Explanations of Unlabeled Data: Why You Should Average
- Stochastic Hyperparameter Optimization through Hypernetworks
- Semi-supervised Deep Kernel Learning: Regression with Unlabeled Data by Minimizing Predictive Variance
- Not All Unlabeled Data are Equal: Learning to Weight Data in Semi-supervised Learning
- SwitchOut: an Efficient Data Augmentation Algorithm for Neural Machine Translation
- Learning to learn via Self-Critique