Adaptive Anchor Label Propagation for Transductive Few-Shot Learning

Michalis Lazarou*, Yannis Avrithis, Guangyu Ren, Tania Stathaki

*Corresponding author for this work

Research output: Chapter in Book or Report/Conference proceedingConference Proceedingpeer-review

Abstract

Few-shot learning addresses the issue of classifying images using limited labeled data. Exploiting unlabeled data through the use of transductive inference methods such as label propagation has been shown to improve the performance of few-shot learning significantly. Label propagation infers pseudo-labels for unlabeled data by utilizing a constructed graph that exploits the underlying manifold structure of the data. However, a limitation of the existing label propagation approaches is that the positions of all data points are fixed and might be sub-optimal so that the algorithm is not as effective as possible. In this work, we propose a novel algorithm that adapts the feature embeddings of the labeled data by minimizing a differentiable loss function optimizing their positions in the manifold in the process. Our novel algorithm, Adaptive Anchor Label Propagation, outperforms the standard label propagation algorithm by as much as 7% and 2% in the 1-shot and 5-shot settings respectively. We provide experimental results highlighting the merits of our algorithm on four widely used few-shot benchmark datasets, namely miniImageNet, tieredImageNet, CUB and CIFAR-FS and two commonly used backbones, ResNet12 and WideResNet-28-10. The source code can be found at https://github.com/MichalisLazarou/A2LP.

Original languageEnglish
Title of host publication2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings
PublisherIEEE Computer Society
Pages331-335
Number of pages5
ISBN (Electronic)9781728198354
DOIs
Publication statusPublished - 2023
Externally publishedYes
Event30th IEEE International Conference on Image Processing, ICIP 2023 - Kuala Lumpur, Malaysia
Duration: 8 Oct 202311 Oct 2023

Publication series

NameProceedings - International Conference on Image Processing, ICIP
ISSN (Print)1522-4880

Conference

Conference30th IEEE International Conference on Image Processing, ICIP 2023
Country/TerritoryMalaysia
CityKuala Lumpur
Period8/10/2311/10/23

Keywords

  • few-shot learning
  • label propagation
  • transductive inference

Fingerprint

Dive into the research topics of 'Adaptive Anchor Label Propagation for Transductive Few-Shot Learning'. Together they form a unique fingerprint.

Cite this