Adaptive Subspaces for Few-Shot Learning
Object recognition requires a generalization capability to avoid overfitting, especially when the samples are extremely few. Generalization from limited samples, usually studied under the umbrella of meta-learning, equips learning techniques with the ability to adapt quickly in dynamical environments and proves to be an essential aspect of life long learning. In this paper, we provide a framework for few-shot learning by introducing dynamic classifiers that are constructed from few samples. A subspace method is exploited as the central block of a dynamic classifier. We will empirically show that such modelling leads to robustness against perturbations (e.g., outliers) and yields competitive results on the task of supervised and semi-supervised few-shot classification. We also develop a discriminative form which can boost the accuracy even further. Our code is available at https://github.com/chrysts/dsn_fewshot
PDF AbstractCode
Results from the Paper
Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Benchmark |
---|---|---|---|---|---|---|
Few-Shot Image Classification | CIFAR-FS 5-way (1-shot) | Adaptive Subspace Network | Accuracy | 78 | # 14 | |
Few-Shot Image Classification | CIFAR-FS 5-way (5-shot) | Adaptive Subspace Network | Accuracy | 87.3 | # 23 | |
Few-Shot Image Classification | Mini-Imagenet 5-way (1-shot) | Adaptive Subspace Network | Accuracy | 67.09 | # 45 | |
Few-Shot Image Classification | Mini-Imagenet 5-way (5-shot) | Adaptive Subspace Network | Accuracy | 81.65 | # 40 | |
Few-Shot Image Classification | Tiered ImageNet 5-way (1-shot) | Adaptive Subspace Network | Accuracy | 68.44 | # 37 | |
Few-Shot Image Classification | Tiered ImageNet 5-way (5-shot) | Adaptive Subspace Network | Accuracy | 83.32 | # 34 |