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

TL;DR

This paper empirically investigates the perceived incompatibility between label smoothing and knowledge distillation, analyzing how label smoothing impacts information retention and its effectiveness across various tasks.

Contribution

It provides a detailed analysis and new metrics to measure information loss due to label smoothing, challenging the notion of their inherent incompatibility.

Findings

Label smoothing erases relative information between classes.

The incompatibility is context-dependent and not absolute.

Certain circumstances exist where label smoothing remains effective.

Abstract

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.