MixMatch swMATH ID: 41280 Software Authors: David Berthelot, Nicholas Carlini, Ian Goodfellow, Nicolas Papernot, Avital Oliver, Colin Raffel Description: Mixmatch: A holistic approach to semi-supervised learning. Semi-supervised learning has proven to be a powerful paradigm for leveraging unlabeled data to mitigate the reliance on large labeled datasets. In this work, we unify the current dominant approaches for semi-supervised learning to produce a new algorithm, MixMatch, that works by guessing low-entropy labels for data-augmented unlabeled examples and mixing labeled and unlabeled data using MixUp. We show that MixMatch obtains state-of-the-art results by a large margin across many datasets and labeled data amounts. For example, on CIFAR-10 with 250 labels, we reduce error rate by a factor of 4 (from 38 Homepage: https://arxiv.org/abs/1905.02249 Source Code: https://github.com/google-research/mixmatch Related Software: CIFAR; PyTorch; Python; SGDR; ImageNet; ReMixMatch; FixMatch; Scikit; mixup; RandAugment; FlexMatch; SemiBoost; S4L; Adam; t-SNE; TensorFlow; LAMDA-SSL; AutoAugment; AugMix; TADAM Cited in: 7 Publications all top 5 Cited by 22 Authors 1 Ahfock, Daniel C. 1 Chen, Songcan 1 Cui, Junbiao 1 Du, Qiang 1 Fan, Ya 1 Harchaoui, Zaid 1 Hoos, Holger H. 1 Huang, Shengjun 1 Jones, Corinne 1 Lang, Rongling 1 Li, Shao-Yuan 1 Liang, Jiye 1 Liu, Guodong 1 Liu, Guoliang 1 McLachlan, Geoffrey John 1 Roulet, Vincent 1 Shi, Ye 1 Sun, WaiChing 1 van Engelen, Jesper E. 1 Wang, Jie 1 Wang, Kun 1 Wei, Wei Cited in 4 Serials 3 Machine Learning 2 Statistics and Computing 1 Computer Methods in Applied Mechanics and Engineering 1 Applied Mathematics and Computation Cited in 5 Fields 5 Statistics (62-XX) 5 Computer science (68-XX) 1 Combinatorics (05-XX) 1 Mechanics of deformable solids (74-XX) 1 Game theory, economics, finance, and other social and behavioral sciences (91-XX) Citations by Year