Discriminative Few-Shot Learning Based on Directional Statistics

Abstract

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.

References in corpus (6)

• Prototypical Networks for Few-shot Learning
• Theoretical Models of Learning to Learn
• A Simple Neural Attentive Meta-Learner
• Gradient-Based Meta-Learning with Learned Layerwise Metric and Subspace
• von Mises-Fisher Mixture Model-based Deep learning: Application to Face Verification
• Von Mises-Fisher Loss for Training Sequence to Sequence Models with Continuous Outputs