Pith. sign in

REVIEW 7 cited by

Dropout as a Bayesian Approximation: Appendix

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1506.02157 v5 pith:C3XFMGLE submitted 2015-06-06 stat.ML

Yarin Gal , Zoubin Ghahramani This is my paper · ORCID
classification stat.ML
keywords bayesiandropoutapproximationdeeplearningallowsappendixinterpretation
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

We show that a neural network with arbitrary depth and non-linearities, with dropout applied before every weight layer, is mathematically equivalent to an approximation to a well known Bayesian model. This interpretation might offer an explanation to some of dropout's key properties, such as its robustness to over-fitting. Our interpretation allows us to reason about uncertainty in deep learning, and allows the introduction of the Bayesian machinery into existing deep learning frameworks in a principled way. This document is an appendix for the main paper "Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning" by Gal and Ghahramani, 2015.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Solar image denoising with convolutional neural networks

    astro-ph.SR 2019-08 conditional novelty 6.0 of 10

    A U-net trained on noisy image pairs alone, without clean targets, denoises solar Stokes images to about 6e-4 continuum residual, matching clean-target training on synthetic data.

  2. Graph Evidential Learning for Anomaly Detection

    cs.LG 2025-05 conditional novelty 5.0 of 10

    GEL detects anomalous nodes by scoring evidential uncertainty from feature and topology reconstruction, reporting gains on four of five benchmark datasets.

  3. Context-Based Echo State Networks with Prediction Confidence for Human-Robot Shared Control

    cs.RO 2024-11 conditional novelty 5.0 of 10

    A reservoir-computing model, CESN+, adds prediction-interval confidence to trajectory generation and uses it to adaptively share control, reducing human effort in simulation.

  4. PolyGraphPy: A unified Python framework for atomistic simulation and machine learning-driven polymer design

    cond-mat.mtrl-sci 2026-06 unverdicted novelty 4.0 of 10

    PolyGraphPy automates DFTB calculations for datasets of monomers and copolymers, uses Bayesian GNNs for property prediction with uncertainty quantification, and applies SELFIES-GPT and BRICS-based GA for de novo polym...

  5. Gamma-Ray Bursts Calibrated by Using Artificial Neural Networks from the Pantheon+ Sample

    astro-ph.CO 2025-06 conditional novelty 4.0 of 10

    The authors use an ANN+BNN to calibrate the Amati relation of gamma-ray bursts from Pantheon+ supernovae and obtain cosmological constraints consistent with Gaussian process methods, finding weak 1σ evidence for evolv...

  6. A method for classification of data with uncertainty using hypothesis testing

    cs.LG 2025-02 conditional novelty 3.0 of 10

    A two-sided empirical quantile test on a model's score lets a binary classifier flag ambiguous or out-of-distribution inputs as uncertain instead of labeling them.

  7. Bayesian Neural Networks: An Introduction and Survey

    stat.ML 2020-06 unverdicted novelty 1.0 of 10

    A survey introducing Bayesian Neural Networks and comparing approximate inference methods to enable uncertainty quantification in neural network predictions.

Pith tools