Diversity With Cooperation: Ensemble Methods for Few-Shot Classification

@article{Dvornik2019DiversityWC,
  title={Diversity With Cooperation: Ensemble Methods for Few-Shot Classification},
  author={Nikita Dvornik and Cordelia Schmid and Julien Mairal},
  journal={2019 IEEE/CVF International Conference on Computer Vision (ICCV)},
  year={2019},
  pages={3722-3730}
}
  • Nikita Dvornik, Cordelia Schmid, Julien Mairal
  • Published in
    IEEE/CVF International…
    2019
  • Computer Science
  • Few-shot classification consists of learning a predictive model that is able to effectively adapt to a new class, given only a few annotated samples. To solve this challenging problem, meta-learning has become a popular paradigm that advocates the ability to ``learn to adapt''. Recent works have shown, however, that simple learning strategies without meta-learning could be competitive. In this paper, we go a step further and show that by addressing the fundamental high-variance issue of few… CONTINUE READING

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    Figures, Tables, and Topics from this paper.

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 17 CITATIONS

    Optimized Generic Feature Learning for Few-shot Classification across Domains

    VIEW 7 EXCERPTS
    CITES BACKGROUND, METHODS & RESULTS

    Adversarial Feature Hallucination Networks for Few-Shot Learning

    VIEW 2 EXCERPTS
    CITES METHODS
    HIGHLY INFLUENCED

    Meta-Learning across Meta-Tasks for Few-Shot Learning

    VIEW 3 EXCERPTS
    HIGHLY INFLUENCED

    Self-Augmentation: Generalizing Deep Networks to Unseen Classes for Few-Shot Learning

    VIEW 5 EXCERPTS
    CITES BACKGROUND, METHODS & RESULTS
    HIGHLY INFLUENCED

    Few-Shot Few-Shot Learning and the role of Spatial Attention

    VIEW 4 EXCERPTS
    CITES METHODS & RESULTS
    HIGHLY INFLUENCED

    Incremental Few-Shot Object Detection

    VIEW 2 EXCERPTS
    CITES METHODS & BACKGROUND

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 37 REFERENCES

    Optimization as a model for fewshot learning

    • S. Ravi, H. Larochelle
    • In International Conference on Learning Representations (ICLR),
    • 2017
    VIEW 15 EXCERPTS
    HIGHLY INFLUENTIAL

    Distilling the Knowledge in a Neural Network

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Dynamic Few-Shot Visual Learning Without Forgetting

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Few-Shot Image Recognition by Predicting Parameters from Activations

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Prototypical Networks for Few-shot Learning

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL