# DropMax: Adaptive Variational Softmax

arXiv: https://arxiv.org/abs/1712.07834 · HTML: https://mlsanity.com/paper/1712.07834/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Dec 2017.

## Built on by (most cited first, 3 of 3)

- [Compact Neural Graphics Primitives with Learned Hash Probing](https://mlsanity.com/paper/2312.17241/) (2023)
- [FocusNet: Classifying Better by Focusing on Confusing Classes](https://mlsanity.com/paper/2110.07307/) (2021)
- [Ghost Loss to Question the Reliability of Training Data](https://mlsanity.com/paper/2109.01504/) (2021)

## Abstract (arXiv)

We propose DropMax, a stochastic version of softmax classifier which at each iteration drops non-target classes according to dropout probabilities adaptively decided for each instance. Specifically, we overlay binary masking variables over class output probabilities, which are input-adaptively learned via variational inference. This stochastic regularization has an effect of building an ensemble classifier out of exponentially many classifiers with different decision boundaries. Moreover, the learning of dropout rates for non-target classes on each instance allows the classifier to focus more on classification against the most confusing classes. We validate our model on multiple public datasets for classification, on which it obtains significantly improved accuracy over the regular softmax classifier and other baselines. Further analysis of the learned dropout probabilities shows that our model indeed selects confusing classes more often when it performs classification.