# Shot in the Dark: Few-Shot Learning with No Base-Class Labels

arXiv: https://arxiv.org/abs/2010.02430 · HTML: https://mlsanity.com/paper/2010.02430/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Oct 2020.

## Built on by (most cited first, 9 of 9)

- [STUNT: Few-shot Tabular Learning with Self-generated Tasks from Unlabeled Tables](https://mlsanity.com/paper/2303.00918/) (2023)
- [Self-Supervision Can Be a Good Few-Shot Learner](https://mlsanity.com/paper/2207.09176/) (2022)
- [Label Hallucination for Few-Shot Classification](https://mlsanity.com/paper/2112.03340/) (2021)
- [A Survey of the Self Supervised Learning Mechanisms for Vision Transformers](https://mlsanity.com/paper/2408.17059/) (2024)
- [BECLR: Batch Enhanced Contrastive Few-Shot Learning](https://mlsanity.com/paper/2402.02444/) (2024)
- [TablEye: Seeing small Tables through the Lens of Images](https://mlsanity.com/paper/2307.02491/) (2023)
- [ProMi: An Efficient Prototype-Mixture Baseline for Few-Shot Segmentation with Bounding-Box Annotations](https://mlsanity.com/paper/2505.12547/) (2025)
- [Trip-ROMA: Self-Supervised Learning with Triplets and Random Mappings](https://mlsanity.com/paper/2107.10419/) (2021)
- [Semi-Supervised SAR ATR Framework with Transductive Auxiliary Segmentation](https://mlsanity.com/paper/2308.16633/) (2023)

## Abstract (arXiv)

Few-shot learning aims to build classifiers for new classes from a small number of labeled examples and is commonly facilitated by access to examples from a distinct set of 'base classes'. The difference in data distribution between the test set (novel classes) and the base classes used to learn an inductive bias often results in poor generalization on the novel classes. To alleviate problems caused by the distribution shift, previous research has explored the use of unlabeled examples from the novel classes, in addition to labeled examples of the base classes, which is known as the transductive setting. In this work, we show that, surprisingly, off-the-shelf self-supervised learning outperforms transductive few-shot methods by 3.9% for 5-shot accuracy on miniImageNet without using any base class labels. This motivates us to examine more carefully the role of features learned through self-supervision in few-shot learning. Comprehensive experiments are conducted to compare the transferability, robustness, efficiency, and the complementarity of supervised and self-supervised features.