Oct 9, 2021 · We present a novel mutual learning framework with transductive learning, which aims at iteratively updating the class prototypes and feature extractor.

(3) Our mutual learning framework significantly improves the per- formance of few-shot bioacoustic event detection over the state-of-the-art methods. 2.

We use transductive inference for few shot learning, which maximizes the mutual information between the query features and their label predictions for a given ...

This work proposes to update class prototypes with transductive inference to make the class prototypes as close to the true class center as possible, ...

More specifically, we propose to update class prototypes with transductive inference to make the class prototypes as close to the true class center as possible.

TI-ML [17, 22] is the state-of-art transductive inference mutual learning framework with extra data augmentation in few-shot bioacoustic event detection. As the ...

In this paper, we propose a meta- learning framework for few-shot bioacoustic event detection chal- lenge. Maximizing inter-class distance and minimizing intra- ...

نویسندگان مشترک ; A mutual learning framework for few-shot sound event detection‏. D Yang, H Wang, Y Zou, Z Ye, W Wang‏. ICASSP 2022-2022 IEEE International ...

A MUTUAL LEARNING FRAMEWORK FOR FEW-SHOT SOUND EVENT DETECTION ; Session: Detection and Classification of Acoustic Scenes and Events VI: Events ; Track: Audio and ...

Mar 17, 2024 · [16] Dongchao Yang, Helin Wang, “A mutual learning framework for few-shot sound event detection,” in IEEE. International Conference on ...