Future-Shot: Few-Shot Learning to tackle new labels on high-dimensional classification problems
Nenhuma Miniatura disponível
Data
2024-02-23
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal de Goiás
Resumo
This thesis introduces a novel approach to address high-dimensional multiclass classification challenges, particularly in dynamic environments where new classes emerge. Named Future-Shot, the method employs metric learning, specifically triplet learning, to train a model capable of generating embeddings for both data points and classes within a shared vector space. This facilitates efficient similarity comparisons using techniques like k-nearest neighbors (\acrshort{knn}), enabling seamless integration of new classes without extensive retraining. Tested on lab-of-origin prediction tasks using the Addgene dataset, Future-Shot achieves top-10 accuracy of $90.39\%$, surpassing existing methods. Notably, in few-shot learning scenarios, it achieves an average top-10 accuracy of $81.2\%$ with just $30\%$ of the data for new classes, demonstrating robustness and efficiency in adapting to evolving class structures
Descrição
Citação
CAMARGO, F. H. F. Future-Shot: Few-Shot Learning to tackle new labels on high-dimensional classification problems. 2024. 75 f. Tese (Doutorado em Ciência da Computação) - Instituto de Informática, Universidade Federal de Goiás, Goiânia, 2024.