A Bayesian encourages dropout

Abstract

Dropout is one of the key techniques to prevent the learning from overfitting. It is explained that dropout works as a kind of modified L2 regularization. Here, we shed light on the dropout from Bayesian standpoint. Bayesian interpretation enables us to optimize the dropout rate, which is beneficial for learning of weight parameters and prediction after learning. The experiment result also encourages the optimization of the dropout.

References in corpus (3)

• Improving neural networks by preventing co-adaptation of feature detectors
• Dropout Training as Adaptive Regularization
• Spatially-sparse convolutional neural networks

Cited by in corpus (14)

• Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning
• Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles
• Variational Dropout and the Local Reparameterization Trick
• Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning
• Regularization for Deep Learning: A Taxonomy
• Advanced Dropout: A Model-free Methodology for Bayesian Dropout Optimization
• Dropout-based Active Learning for Regression
• Dropout Feature Ranking for Deep Learning Models
• The Implicit and Explicit Regularization Effects of Dropout
• Prediction intervals for Deep Neural Networks
• Structured Dropout Variational Inference for Bayesian Neural Networks
• BayesGrad: Explaining Predictions of Graph Convolutional Networks
• Dropout Training for SVMs with Data Augmentation
• CODA: Constructivism Learning for Instance-Dependent Dropout Architecture Construction