Leveraging the Feature Distribution in Transfer-based Few-Shot Learning
By Yuqing Hu, Vincent Gripon, Stéphane Pateux. 2020
189 citations recorded by the source corpus.
Few-shot classification is a challenging problem due to the uncertainty caused by using few labelled samples. In the past few years, transfer-based methods have proved to achieve the best performance, thanks to well-thought-out backbone architectures combined with efficient postprocessing steps. Following this vein, in this paper we propose a transfer-based novel method that builds on two steps: 1) preprocessing the feature vectors so that they become closer to Gaussian-like distributions, and 2) leveraging this preprocessing using an optimal-transport inspired algorithm. Using standardized vision benchmarks, we prove the ability of the proposed methodology to achieve state-of-the-art accuracy with various datasets, backbone architectures and few-shot settings.
Subfields: Artificial Intelligence, Computational Mechanics, Meta & Continual Learning
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Lucas B. V. de Amorim
- Katja Hofmann
- Daniela Massiceti
- Rafael M. O. Cruz
- George D. C. Cavalcanti
- Edward Cutrell
- Simone Stumpf
- Luisa M. Zintgraf
In Meta & Continual Learning
Citations in the corpus
- ORBIT: A Real-World Few-Shot Dataset for Teachable Object Recognition (2021, background): “While these datasets have catalyzed research in few-shot learning, state-of-the-art performance is now relatively saturated and leaves reduced scope for algorithmic innovation [16, 4, 33].”
- The choice of scaling technique matters for classification performance (2022, background): “It is able to change the shape of the original attribute distribution, which can be particularly useful as there has been reported evidence that transforming attributes so that they follow a Gaussian-like distribution may benefit classification performance [10].”
- A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities (2022, unknown): “54% ✓ PT-MAP [234] ICANN’20 Metric WRN ✓ ✓ 82.”