Logarithm-transform aided Gaussian sampling for Few-Shot Learning

Vaibhav Ganatra

Keywords: Few-Shot Learning, Gaussian Sampling, Representation Learning

Abstract: Few-shot classification has recently witnessed the rise of representation learning being utilised for models to adapt to new classes using only a few training examples. Therefore, the properties of the representations, such as their underlying probability distributions, assume vital importance. Representations sampled from gaussian distributions have been used in recent works, [19] to train classifiers for few-shot classification. These methods rely on transforming experimental data to approximate gaussian distributions for their functioning. In this paper, we propose a novel gaussian transform, that outperforms existing methods for transforming experimental data into gaussian-like distributions. We then utilise this novel transformation for few-shot image classification and show significant gains in performance, while simultaneously reducing computation.

Submission Number: 24