Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Unsupervised Few-shot Learning via Self-supervised Training

About

Learning from limited exemplars (few-shot learning) is a fundamental, unsolved problem that has been laboriously explored in the machine learning community. However, current few-shot learners are mostly supervised and rely heavily on a large amount of labeled examples. Unsupervised learning is a more natural procedure for cognitive mammals and has produced promising results in many machine learning tasks. In the current study, we develop a method to learn an unsupervised few-shot learner via self-supervised training (UFLST), which can effectively generalize to novel but related classes. The proposed model consists of two alternate processes, progressive clustering and episodic training. The former generates pseudo-labeled training examples for constructing episodic tasks; and the later trains the few-shot learner using the generated episodic tasks which further optimizes the feature representations of data. The two processes facilitate with each other, and eventually produce a high quality few-shot learner. Using the benchmark dataset Omniglot and Mini-ImageNet, we show that our model outperforms other unsupervised few-shot learning methods. Using the benchmark dataset Market1501, we further demonstrate the feasibility of our model to a real-world application on person re-identification.

Zilong Ji, Xiaolong Zou, Tiejun Huang, Si Wu• 2019

Related benchmarks

TaskDatasetResultRank
5-way ClassificationminiImageNet (test)--
231
Image ClassificationMiniImagenet
Accuracy45.03
206
Few-shot classificationMini-Imagenet (test)--
113
Few-shot classificationOmniglot (test)
Accuracy99.19
109
5-Shot 5-Way ClassificationminiImageNet (test)
Accuracy45.03
36
N-way K-shot classificationMini-Imagenet (test)
Accuracy (5-way, 1-shot)33.77
26
5-way 1-shot Image ClassificationminiImageNet standard (test)
Accuracy33.77
12
N-way K-shot Image ClassificationOmniglot (test)
Accuracy (5,1)97.03
11
Showing 8 of 8 rows

Other info

Follow for update