Episodic Fine-Tuning Prototypical Networks for Optimization-Based Few-Shot Learning: Application to Audio Classification

The Prototypical Network (ProtoNet) has emerged as a popular choice in Few-shot Learning (FSL) scenarios due to its remarkable performance and straightforward implementation. Building upon such success, we first p ropose a simple (yet n ovel) method to fine-tune a P rotoNet o n t he (labeled) s upport s et o f t he test episode of a <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$C$</tex>-way-<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$K$</tex>-shot test episode (without using the query set which is only used for evaluation). We then propose an algorithmic framework that combines ProtoNet with optimization-based FSL algorithms (MAML and Meta-Curvature) to work with such a fine-tuning m ethod. Since optimization-based algorithms endow the target learner model with the ability to fast adaption to only a few samples, we utilize ProtoNet as the target model to enhance its fine-tuning performance with the help of a specifically designed episodic fine-tuning s trategy. The experimental results confirm that our proposed models, MAML-Proto and MC-Proto, combined with our unique fine-tuning m ethod, o utperform regular P rotoNet b y a large margin in few-shot audio classification t asks on t he ESC-50 and Speech Commands v2 datasets. We note that although we have only applied our model to the audio domain, it is a general method and can be easily extended to other domains.