Few-shot Learning as Cluster-induced Voronoi Diagrams: A Geometric Approach
Not on the map. Why?
Abstract (arXiv)
Few-shot learning (FSL) is the process of rapid generalization from abundant base samples to inadequate novel samples. Despite extensive research in recent years, FSL is still not yet able to generate satisfactory solutions for a wide range of real-world applications. To confront this challenge, we study the FSL problem from a geometric point of view in this paper. One observation is that the widely embraced ProtoNet model is essentially a Voronoi Diagram (VD) in the feature space. We retrofit it by making use of a recent advance in computational geometry called Cluster-induced Voronoi Diagram (CIVD). Starting from the simplest nearest neighbor model, CIVD gradually incorporates cluster-to-point and then cluster-to-cluster relationships for space subdivision, which is used to improve the accuracy and robustness at multiple stages of FSL. Specifically, we use CIVD (1) to integrate parametric and nonparametric few-shot classifiers; (2) to combine feature representation and surrogate representation; (3) and to leverage feature-level, transformation-level, and geometry-level heterogeneities for a better ensemble. Our CIVD-based workflow enables us to achieve new state-of-the-art results on mini-ImageNet, CUB, and tiered-ImagenNet datasets, with ~2%-5% improvements upon the next best. To summarize, CIVD provides a mathematically elegant and geometrically interpretable framework that compensates for extreme data insufficiency, prevents overfitting, and allows for fast geometric ensemble for thousands of individual VD. These together make FSL stronger.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 38 of its 66 references that are in the map. First posted on arXiv 5 Feb 2022.
Builds on
38 of 66 references.
References · 38
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
- 2016Densely Connected Convolutional Networkscited by 6,691
- 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
- 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
- 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
- 2016Wide Residual Networkscited by 3,057
- 2017Prototypical Networks for Few-shot Learningcited by 3,001
- 2016Matching Networks for One Shot Learningcited by 2,246
- 2019A Closer Look at Few-shot Classificationcited by 679
- 2018Meta-Learning with Latent Embedding Optimizationcited by 476
- 2019Meta-Learning with Differentiable Convex Optimizationcited by 461
26 more
- 2018Meta-Learning for Semi-Supervised Few-Shot Classificationcited by 456
- 2017Meta-SGD: Learning to Learn Quickly for Few-Shot Learningcited by 377
- 2016On the Expressive Power of Deep Neural Networkscited by 302
- 2018Learning to Propagate Labels: Transductive Propagation Network for Few-shot Learningcited by 208
- 2019SimpleShot: Revisiting Nearest-Neighbor Classification for Few-Shot Learningcited by 194
- 2021Free Lunch for Few-shot Learning: Distribution Calibrationcited by 165
- 2019Charting the Right Manifold: Manifold Mixup for Few-shot Learningcited by 135
- 2020Negative Margin Matters: Understanding Margin in Few-shot Classificationcited by 129
- 2020Adversarial Feature Hallucination Networks for Few-Shot Learningcited by 110
- 2018Delta-encoder: an effective sample synthesis method for few-shot object recognitioncited by 89
- 2020Boosting Few-Shot Learning With Adaptive Margin Losscited by 83
- 2020Leveraging the Feature Distribution in Transfer-based Few-Shot Learningcited by 72
- 2019Diversity with Cooperation: Ensemble Methods for Few-Shot Classificationcited by 70
- 2018Multi-level Semantic Feature Augmentation for One-shot Learningcited by 62
- 2019The Geometry of Deep Networks: Power Diagram Subdivisioncited by 53
- 2020Prototype Completion with Primitive Knowledge for Few-Shot Learningcited by 50
- 2021Bridging Multi-Task Learning and Meta-Learning: Towards Efficient Training and Effective Adaptationcited by 50
- 2020DeepEMD: Differentiable Earth Mover's Distance for Few-Shot Learningcited by 49
- 2019An Ensemble of Epoch-wise Empirical Bayes for Few-shot Learningcited by 38
- 2020Data Augmentation for Meta-Learningcited by 38
- 2019Neural Style Transfer Improves 3D Cardiovascular MR Image Segmentation on Inconsistent Datacited by 18
- 2020Task Augmentation by Rotating for Meta-Learningcited by 18
- 2020Learning to Select Base Classes for Few-shot Classificationcited by 11
- 2021Improving Uncertainty Calibration of Deep Neural Networks via Truth Discovery and Geometric Optimizationcited by 11
- 2020Max-Affine Spline Insights into Deep Generative Networkscited by 10
- 2020Impact of base dataset design on few-shot image classificationcited by 9
Built on by
Citing papers per year: 2023: 1; 2024: 2.
3 papers.