Dropout: A Simple Way to Prevent Neural Networks from Overfitting ​

← Back to topic

Authors: Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, Ruslan Salakhutdinov
Year: 2014
Journal: JMLR
DOI: 10.5555/2627435.2670313
Publisher: https://jmlr.org/papers/v15/srivastava14a.html

Keywords: dropout, regularization, deep learning

Abstract ​

Dropout is a technique for addressing overfitting in deep neural networks by randomly dropping units during training.

Cite this paper ​

@misc{dropout2014,
  title  = {Dropout: A Simple Way to Prevent Neural Networks from Overfitting},
  author = {Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, Ruslan Salakhutdinov},
  year   = {2014},
  journal = {JMLR},
  doi    = {10.5555/2627435.2670313},
  url    = {https://doi.org/10.5555/2627435.2670313},
}

Source files ​

• metadata.json
• paper.bib