PofoliaShared via Pofolia

arXiv (Cornell University)· 2019· Preprint

When Does Label Smoothing Help?

Rafael Rios Müller, Simon Kornblith, Geoffrey E. Hinton

Short summary

Label smoothing improves neural network generalization and calibration, but hinders knowledge distillation when used in teacher models.

AI-generated from the title and abstract; the full text is not read.

TakeawaysIn the app
Key pointsIn the app
Ask the paperIn the app

The rest is in the Pofolia app

Takeaways, key points and questions to the paper; new summaries every day for your field. Free.

Sign in on the web to open

Field: Signal Processing

Signal ProcessingComputer Science