DP-SSL: Towards Robust Semi-supervised Learning with A Few Labeled Samples

The scarcity of labeled data is a critical obstacle to deep learning.-supervised learning (SSL) provides a promising way to leverage unlabeled by pseudo labels. However, when the size of labeled data is very small(say a few labeled samples per class), SSL performs poorly and unstably, due to the low quality of learned pseudo labels. In this paper, we a new SSL method called DP-SSL that adopts an innovative data (DP) scheme to generate probabilistic labels for unlabeled data. from existing DP methods that rely on human experts to provide labeling functions (LFs), we develop a multiple-choice learning~(MCL) approach to automatically generate LFs from scratch in SSL style. With noisy labels produced by the LFs, we design a label model to resolve the and overlap among the noisy labels, and finally infer probabilistic for unlabeled samples. Extensive experiments on four standard SSL show that DP-SSL can provide reliable labels for unlabeled data and better classification performance on test sets than existing SSL, especially when only a small number of labeled samples are available., for CIFAR-10 with only 40 labeled samples, DP-SSL achieves 93.82% accuracy on unlabeled data and 93.46% classification accuracy on data, which are higher than the SOTA results.