# Max-Pooling Dropout for Regularization of Convolutional Neural Networks

arXiv: https://arxiv.org/abs/1512.01400 · HTML: https://mlsanity.com/paper/1512.01400/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Dec 2015.

## Built on by (most cited first, 11 of 11)

- [DropoutGS: Dropping Out Gaussians for Better Sparse-view Rendering](https://mlsanity.com/paper/2504.09491/) (2025)
- [Triangular Bidword Generation for Sponsored Search Auction](https://mlsanity.com/paper/2101.11349/) (2021)
- [A General Framework for Robust G-Invariance in G-Equivariant Networks](https://mlsanity.com/paper/2310.18564/) (2023)
- [Deep Neural Networks Inspired by Differential Equations](https://mlsanity.com/paper/2510.09685/) (2025)
- [FSER: Deep Convolutional Neural Networks for Speech Emotion Recognition](https://mlsanity.com/paper/2109.07916/) (2021)
- [AI-enabled Sound Pattern Recognition on Asthma Medication Adherence: Evaluation with the RDA Benchmark Suite](https://mlsanity.com/paper/2205.15360/) (2022)
- [Patient-specific modelling, simulation and real-time processing for respiratory diseases](https://mlsanity.com/paper/2207.01082/) (2022)
- [Automated Assessment of Transthoracic Echocardiogram Image Quality Using Deep Neural Networks](https://mlsanity.com/paper/2209.00976/) (2022)
- [Virtual Underwater Datasets for Autonomous Inspections](https://mlsanity.com/paper/2209.06013/) (2022)
- [Performance Comparison of Simple Transformer and Res-CNN-BiLSTM for Cyberbullying Classification](https://mlsanity.com/paper/2206.02206/) (2022)
- [Based on What We Can Control Artificial Neural Networks](https://mlsanity.com/paper/2310.05692/) (2023)

## Abstract (arXiv)

Recently, dropout has seen increasing use in deep learning. For deep convolutional neural networks, dropout is known to work well in fully-connected layers. However, its effect in pooling layers is still not clear. This paper demonstrates that max-pooling dropout is equivalent to randomly picking activation based on a multinomial distribution at training time. In light of this insight, we advocate employing our proposed probabilistic weighted pooling, instead of commonly used max-pooling, to act as model averaging at test time. Empirical evidence validates the superiority of probabilistic weighted pooling. We also compare max-pooling dropout and stochastic pooling, both of which introduce stochasticity based on multinomial distributions at pooling stage.