Survey on Meta Learning Algorithms for Few Shot Learning

Abstract

In Meta learning auto learning algorithms are applied to machine learning experiments. The Meta learning is trying to solve problem of learning to learn as there is a significant lack in data as well as experts. There are many novel approaches developed in field of meta learning in past few years. This paper is summary of ongoing research in field of meta learning. It describes current trends and development in field of meta learning and with tactic knowledge how the meta learning can be applied to achieve few shot learning. It is believe that the Meta learning will perform well to overcome the challenges of few shot learning.