Dropout Inference in Bayesian Neural Networks with Alpha-divergences

Abstract

To obtain uncertainty estimates with real-world Bayesian deep learning models, practical inference approximations are needed. Dropout variational inference (VI) for example has been used for machine vision and medical applications, but VI can severely underestimates model uncertainty. Alpha-divergences are alternative divergences to VI's KL objective, which are able to avoid VI's uncertainty underestimation. But these are hard to use in practice: existing techniques can only use Gaussian approximating distributions, and require existing models to be changed radically, thus are of limited use for practitioners. We propose a re-parametrisation of the alpha-divergence objectives, deriving a simple inference technique which, together with dropout, can be easily implemented with existing models by simply changing the loss of the model. We demonstrate improved uncertainty estimates and accuracy compared to VI in dropout networks. We study our model's epistemic uncertainty far away from the data using adversarial images, showing that these can be distinguished from non-adversarial images by examining our model's uncertainty.

References in corpus (9)

• Explaining and Harnessing Adversarial Examples
• Improving neural networks by preventing co-adaptation of feature detectors
• Rényi Divergence and Kullback-Leibler Divergence
• Expectation Propagation for approximate Bayesian inference
• Bayesian Convolutional Neural Networks with Bernoulli Approximate Variational Inference
• Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networks
• Deep Gaussian Processes for Regression using Approximate Expectation Propagation
• Rényi Divergence Variational Inference
• Stochastic Expectation Propagation

Cited by in corpus (45)

• Hands-on Bayesian Neural Networks -- a Tutorial for Deep Learning Users
• Quantifying total uncertainty in physics-informed neural networks for solving forward and inverse stochastic problems
• Deep and Confident Prediction for Time Series at Uber
• Bayesian Uncertainty Estimation for Batch Normalized Deep Networks
• Adversarial Attack Vulnerability of Medical Image Analysis Systems: Unexplored Factors
• Adversarial Examples, Uncertainty, and Transfer Testing Robustness in Gaussian Process Hybrid Deep Networks
• Model-Predictive Policy Learning with Uncertainty Regularization for Driving in Dense Traffic
• Advanced Dropout: A Model-free Methodology for Bayesian Dropout Optimization
• Boosting Adversarial Attacks with Momentum
• Survey of Dropout Methods for Deep Neural Networks
• Variational Inference MPC for Bayesian Model-based Reinforcement Learning
• Variational Inference based on Robust Divergences
• Cognitive simulation models for inertial confinement fusion: Combining simulation and experimental data
• Information Aware Max-Norm Dirichlet Networks for Predictive Uncertainty Estimation
• Sufficient Conditions for Idealised Models to Have No Adversarial Examples: a Theoretical and Empirical Study with Bayesian Neural Networks
• Robustness of Bayesian Neural Networks to Gradient-Based Attacks
• Are Generative Classifiers More Robust to Adversarial Attacks?
• Understanding the Decision Boundary of Deep Neural Networks: An Empirical Study
• Active Learning for Deep Detection Neural Networks
• From Deterministic to Generative: Multi-Modal Stochastic RNNs for Video Captioning
• Sampling-Free Variational Inference of Bayesian Neural Networks by Variance Backpropagation
• Building robust classifiers through generation of confident out of distribution examples
• Towards Robust Detection of Adversarial Examples
• The Hidden Uncertainty in a Neural Networks Activations
• Bayesian Neural Networks
• Variational Implicit Processes
• Bayesian Adversarial Spheres: Bayesian Inference and Adversarial Examples in a Noiseless Setting
• Parametric Gaussian Process Regressors
• On the use of uncertainty in classifying Aedes Albopictus mosquitoes
• How Wrong Am I? - Studying Adversarial Examples and their Impact on Uncertainty in Gaussian Process Machine Learning Models
• Stochastic Maximum Likelihood Optimization via Hypernetworks
• Bayesian regression explains how human participants handle parameter uncertainty
• Bayesian Optimization Meets Laplace Approximation for Robotic Introspection
• Selective Dyna-style Planning Under Limited Model Capacity
• Bayesian posterior approximation with stochastic ensembles
• Uncertainty-driven ensembles of deep architectures for multiclass classification. Application to COVID-19 diagnosis in chest X-ray images
• Reliable Uncertainties for Bayesian Neural Networks using Alpha-divergences
• Stochastic-Shield: A Probabilistic Approach Towards Training-Free Adversarial Defense in Quantized CNNs
• Requirements for Developing Robust Neural Networks
• SINVAD: Search-based Image Space Navigation for DNN Image Classifier Test Input Generation
• Multi-person Pose Tracking using Sequential Monte Carlo with Probabilistic Neural Pose Predictor
• Contributions to Large Scale Bayesian Inference and Adversarial Machine Learning
• An Adaptive Empirical Bayesian Method for Sparse Deep Learning
• Quantal synaptic dilution enhances sparse encoding and dropout regularisation in deep networks
• Robustness Against Outliers For Deep Neural Networks By Gradient Conjugate Priors