Discriminative Few-Shot Learning Based on Directional Statistics
This addresses the problem of limited training data in few-shot classification for computer vision applications, representing an incremental improvement over existing metric-based methods.
The paper tackles few-shot learning by proposing a novel algorithm that generates class representatives using a mixture of von Mises-Fisher distributions, which outperforms comparable methods on miniImageNet and tieredImageNet datasets.
Metric-based few-shot learning methods try to overcome the difficulty due to the lack of training examples by learning embedding to make comparison easy. We propose a novel algorithm to generate class representatives for few-shot classification tasks. As a probabilistic model for learned features of inputs, we consider a mixture of von Mises-Fisher distributions which is known to be more expressive than Gaussian in a high dimensional space. Then, from a discriminative classifier perspective, we get a better class representative considering inter-class correlation which has not been addressed by conventional few-shot learning algorithms. We apply our method to \emph{mini}ImageNet and \emph{tiered}ImageNet datasets, and show that the proposed approach outperforms other comparable methods in few-shot classification tasks.