Semi-Supervised Learning (SSL) algorithms have shown great potential in training regimes when access to labeled data is scarce but access to unlabeled data is plentiful. However, our experiments illustrate several shortcomings that prior SSL algorithms suffer from. In particular, poor performance when unlabeled and labeled data distributions differ. To address these observations, we develop RealMix, which achieves state-of-the-art results on standard benchmark datasets across different labeled and unlabeled set sizes while overcoming the aforementioned challenges. Notably, RealMix achieves an error rate of 9.79% on CIFAR10 with 250 labels, and is the only SSL method tested able to surpass baseline performance when there is significant mismatch in the labeled and unlabeled data distributions. RealMix demonstrates how SSL can be used in real world situations with limited access to both data and compute and guides further research in SSL with practical applicability in mind.

No abstract

No abstract

No abstract

Purpose of the paper:With the purpose of fostering a continuous improvement policy in the 25 services provided by the University of Oviedo aimed at increasing the quality, effectiveness and efficiency of these services, the Technical Quality Unit has drawn up a specific Quality Plan for Services.The main goals are: -To foster public information on the services provided and on the quality commitments assumed in their provision via the drawing up of Service Charters to users. -To improve the quality and efficiency of the services provided to users through the implementation of an Internal Quality Assurance System. -The obtaining of external accreditations.