# Single Shot MC Dropout Approximation

arXiv: https://arxiv.org/abs/2007.03293 · HTML: https://mlsanity.com/paper/2007.03293/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Jul 2020.

## Built on by (most cited first, 10 of 10)

- [A Survey on Epistemic (Model) Uncertainty in Supervised Learning: Recent Advances and Applications](https://mlsanity.com/paper/2111.01968/) (2021)
- [B-BACN: Bayesian Boundary-Aware Convolutional Network for Crack Characterization](https://mlsanity.com/paper/2302.06827/) (2023)
- [Error-Driven Uncertainty Aware Training](https://mlsanity.com/paper/2405.01205/) (2024)
- [Uncertainty Estimation by Human Perception versus Neural Models](https://mlsanity.com/paper/2506.15850/) (2025)
- [CLUE: Neural Networks Calibration via Learning Uncertainty-Error alignment](https://mlsanity.com/paper/2505.22803/) (2025)
- [Intrinsic uncertainties and where to find them](https://mlsanity.com/paper/2107.02526/) (2021)
- [MP-GELU Bayesian Neural Networks: Moment Propagation by GELU Nonlinearity](https://mlsanity.com/paper/2211.13402/) (2022)
- [Single-shot Bayesian approximation for neural networks](https://mlsanity.com/paper/2308.12785/) (2023)
- [A Saliency-based Clustering Framework for Identifying Aberrant Predictions](https://mlsanity.com/paper/2311.06454/) (2023)
- [MetaErr: Towards Predicting Error Patterns in Deep Neural Networks](https://mlsanity.com/paper/2604.23289/) (2026)

## Abstract (arXiv)

Deep neural networks (DNNs) are known for their high prediction performance, especially in perceptual tasks such as object recognition or autonomous driving. Still, DNNs are prone to yield unreliable predictions when encountering completely new situations without indicating their uncertainty. Bayesian variants of DNNs (BDNNs), such as MC dropout BDNNs, do provide uncertainty measures. However, BDNNs are slow during test time because they rely on a sampling approach. Here we present a single shot MC dropout approximation that preserves the advantages of BDNNs without being slower than a DNN. Our approach is to analytically approximate for each layer in a fully connected network the expected value and the variance of the MC dropout signal. We evaluate our approach on different benchmark datasets and a simulated toy example. We demonstrate that our single shot MC dropout approximation resembles the point estimate and the uncertainty estimate of the predictive distribution that is achieved with an MC approach, while being fast enough for real-time deployments of BDNNs.