Language Modeling
Machine Learning
A regularization technique that softens hard target labels to improve model generalization and prevent overconfidence. This involves mixing target labels with a uniform distribution.
Detailed Explanation
Label Smoothing is a regularization technique in machine learning that replaces hard, one-hot target labels with soft labels by blending them with a uniform distribution. This prevents the model from becoming overconfident in its predictions, encouraging better generalization, reducing overfitting, and improving model calibration by providing softer, more informative target signals during training.
Use Cases
- •Use case: Enhances model calibration and generalization by preventing overconfidence during training in image classification tasks.
Related Terms
Other terms in the Machine Learning category