Interpolation Consistency Training for Semi-Supervised Learning
This method addresses the problem of limited labeled data in classification tasks, offering a simple and efficient solution for semi-supervised learning.
The paper tackles semi-supervised learning by introducing Interpolation Consistency Training (ICT), which encourages consistency in predictions for interpolated unlabeled data, achieving state-of-the-art performance on CIFAR-10 and SVHN datasets.
We introduce Interpolation Consistency Training (ICT), a simple and computation efficient algorithm for training Deep Neural Networks in the semi-supervised learning paradigm. ICT encourages the prediction at an interpolation of unlabeled points to be consistent with the interpolation of the predictions at those points. In classification problems, ICT moves the decision boundary to low-density regions of the data distribution. Our experiments show that ICT achieves state-of-the-art performance when applied to standard neural network architectures on the CIFAR-10 and SVHN benchmark datasets. Our theoretical analysis shows that ICT corresponds to a certain type of data-adaptive regularization with unlabeled points which reduces overfitting to labeled points under high confidence values.