.png){kind=icon}
Return
Deep label embedding learning for classification
.png){kind=icon}
PRE
AI
DOI：10.1016/j.asoc.2024.111925.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The one-hot 0/1 encoding method is the most popularized encoding method of class labels for classification tasks. Despite its simplicity and popularity, it comes with limitations and weaknesses, like failing to capture the inherent uncertainty in data labels, and making classifiers more prone to overfitting. In this paper, these shortcomings are tackled with a framework for learning soft label embeddings. Two variants are proposed: first, a learnable general-class embedding which aims to capture information regarding inter-class similarities, and second, a neural architecture which can be added to any neural classifier and aims to learn inter-instance similarities. The inherent uncertainty in data labels is thus somewhat alleviated, allowing the network to focus on incorrectly classified samples, instead of difficult but correctly classified ones. The experimental study on multiple classification benchmarks of increasing difficulty, using neural networks of varying depth and width, show that the proposed method leads to better classification accuracy, highlighting its ability to generalize to unseen samples.
Keywords:
Label embedding
Soft labels
Class similarities
Instance similarities
Journal
IF:
6.6
Papers:
1.4W
Citations:
4.8W
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSaveThe impact of financial development on the upgrading of China’s export technical sophistication
PREAI
ShareSaveShareSave
ShareSaveShareSave
View more