mlsanityData to 14 Oct 2026

Universum Prescription: Regularization using Unlabeled Data

arXiv 1511.03719 ↗First posted 11 Nov 2015cs.LGcited by 16

Not on the map. Why?

Abstract (arXiv)

This paper shows that simply prescribing "none of the above" labels to unlabeled data has a beneficial regularization effect to supervised learning. We call it universum prescription by the fact that the prescribed labels cannot be one of the supervised labels. In spite of its simplicity, universum prescription obtained competitive results in training deep convolutional networks for CIFAR-10, CIFAR-100, STL-10 and ImageNet datasets. A qualitative justification of these approaches using Rademacher complexity is presented. The effect of a regularization parameter -- probability of sampling from unlabeled data -- is also studied empirically.

Key facts. Cited by 16 papers in the map (data through 14 Oct 2026). First posted on arXiv 11 Nov 2015.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 9; 2022: 3; 2023: 3; 2024: 1.

16 papers.

Sort:

  1. 2021SegmentMeIfYouCan: A Benchmark for Anomaly Segmentationcited by 150
  2. 2021Does Your Dermatology Classifier Know What It Doesn't Know? Detecting the Long-Tail of Unseen Conditionscited by 68
  3. 2021Taxonomy of Machine Learning Safety: A Survey and Primercited by 25
  4. 2021Being a Bit Frequentist Improves Bayesian Neural Networkscited by 15
  5. 2022Correct-by-Construction Runtime Enforcement in AI -- A Surveycited by 12
  6. 2021DOC3-Deep One Class Classification using Contradictionscited by 4
  7. 2021Towards Consistent Predictive Confidence through Fitted Ensemblescited by 4
  8. 2022Class-Aware Universum Inspired Re-Balance Learning for Long-Tailed Recognitioncited by 3
  9. 2024All Beings Are Equal in Open Set Recognitioncited by 3
  10. 2022Universum-inspired Supervised Contrastive Learningcited by 2
  11. 2021Automatic Large Scale Detection of Red Palm Weevil Infestation using Aerial and Street View Imagescited by 1
  12. 2021Universum GANs: Improving GANs through contradictionscited by 1
4 more
  1. 2021One Representation to Rule Them All: Identifying Out-of-Support Examples in Few-shot Learning with Generic Representationscited by 0
  2. 2023On Feature Diversity in Energy-based Modelscited by 0
  3. 2023Defensive Perception: Estimation and Monitoring of Neural Network Performance under Deploymentcited by 0
  4. 2023Preventing Arbitrarily High Confidence on Far-Away Data in Point-Estimated Discriminative Neural Networkscited by 0