paper-with-me

홈 › Papers

Is Label Smoothing Truly Incompatible with Knowledge Distillation: An Empirical Study

2021-04-01 · ICLR 2021 1 · Zhiqiang Shen, Zechun Liu, Dejia Xu, Zitian Chen, Kwang-Ting Cheng, Marios Savvides

This work aims to empirically clarify a recently discovered perspective that label smoothing is incompatible with knowledge distillation. We begin by introducing the motivation behind on how this incompatibility is raised, i.e., label smoothing erases relative information between teacher logits. We provide a novel connection on how label smoothing affects distributions of semantically similar and dissimilar classes. Then we propose a metric to quantitatively measure the degree of erased information in sample's representation. After that, we study its one-sidedness and imperfection of the incompatibility view through massive analyses, visualizations and comprehensive experiments on Image Classification, Binary Networks, and Neural Machine Translation. Finally, we broadly discuss several circumstances wherein label smoothing will indeed lose its effectiveness. Project page: http://zhiqiangshen.com/projects/LS_and_KD/index.html.

📄 PDF Abstract BibTeX arXiv:2104.00676

Code (0)

등록된 구현이 없습니다.

Tasks

image-classificationImage ClassificationKnowledge DistillationMachine TranslationTranslation

Methods 이 논문이 사용한 방법론

Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…

Similar Papers 제목 키워드 기반

Adaptive Label Smoothing with Self-Knowledge in Natural Language Generation

2022-10-22 · Dongkyu Lee, Ka Chun Cheung, Nevin L. Zhang

Overconfidence has been shown to impair generalization and calibration of a neural network. Previous studies remedy this issue by adding a regularization term to a loss function, preventing a model from making a peaked d…

Knowledge DistillationText Generation

Adaptive Label Smoothing with Self-Knowledge

2021-09-29 · Dongkyu Lee, Ka Chun Cheung, Nevin Zhang

Overconfidence has been shown to impair generalization and calibration of a neural network. Previous studies remedy this issue by adding a regularization term to a loss function, preventing a model from making a peaked d…

Knowledge DistillationMachine Translation

Pseudo Knowledge Distillation: Towards Learning Optimal Instance-specific Label Smoothing Regularization

2021-09-29 · Peng Lu, Ahmad Rashid, Ivan Kobyzev, Mehdi Rezagholizadeh 외

Knowledge Distillation (KD) is an algorithm that transfers the knowledge of a trained, typically larger, neural network into another model under training. Although a complete understanding of KD is elusive, a growing bod…

image-classificationImage ClassificationKnowledge DistillationNatural Language Understanding

Distilling Knowledge from Pre-trained Language Models via Text Smoothing

2020-05-08 · Xing Wu, Yibing Liu, Xiangyang Zhou, dianhai yu

This paper studies compressing pre-trained language models, like BERT (Devlin et al.,2019), via teacher-student knowledge distillation. Previous works usually force the student model to strictly mimic the smoothed labels…

Knowledge DistillationLanguage ModelingLanguage Modelling

When Does Label Smoothing Help?

2019-06-06 · NeurIPS 2019 12 · Rafael Müller, Simon Kornblith, Geoffrey Hinton

The generalization and learning speed of a multi-class neural network can often be significantly improved by using soft targets that are a weighted average of the hard targets and the uniform distribution over labels. Sm…

image-classificationImage ClassificationKnowledge Distillationspeech-recognition+2