Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning

Abstract

Deep learning tools have gained tremendous attention in applied machine learning. However such tools for regression and classification do not capture model uncertainty. In comparison, Bayesian models offer a mathematically grounded framework to reason about model uncertainty, but usually come with a prohibitive computational cost. In this paper we develop a new theoretical framework casting dropout training in deep neural networks (NNs) as approximate Bayesian inference in deep Gaussian processes. A direct result of this theory gives us tools to model uncertainty with dropout NNs -- extracting information from existing models that has been thrown away so far. This mitigates the problem of representing uncertainty in deep learning without sacrificing either computational complexity or test accuracy. We perform an extensive study of the properties of dropout's uncertainty. Various network architectures and non-linearities are assessed on tasks of regression and classification, using MNIST as an example. We show a considerable improvement in predictive log-likelihood and RMSE compared to existing state-of-the-art methods, and finish by using dropout's uncertainty in deep reinforcement learning.

12 pages, 6 figures; fixed a mistake with standard error and added a new table with updated results (marked "Update [October 2016]"); Published in ICML 2016

References in corpus (8)

• Adam: A Method for Stochastic Optimization
• Practical Bayesian Optimization of Machine Learning Algorithms
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Weight Uncertainty in Neural Networks
• Compressing Neural Networks with the Hashing Trick
• Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networks
• Variational Bayesian Inference with Stochastic Search
• A Bayesian encourages dropout

Cited by in corpus (325)

• Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentation
• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• Recent Advances and Applications of Deep Learning Methods in Materials Science
• A trans-disciplinary review of deep learning research for water resources scientists
• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• Hands-on Bayesian Neural Networks -- a Tutorial for Deep Learning Users
• Interactive Medical Image Segmentation using Deep Learning with Image-specific Fine-tuning
• A Theoretically Grounded Application of Dropout in Recurrent Neural Networks
• Bayesian Deep Convolutional Encoder-Decoder Networks for Surrogate Modeling and Uncertainty Quantification
• Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networks
• A Survey on Active Learning and Human-in-the-Loop Deep Learning for Medical Image Analysis
• Bayesian Convolutional Neural Networks with Bernoulli Approximate Variational Inference
• Deep Bayesian Active Learning with Image Data
• Quantifying total uncertainty in physics-informed neural networks for solving forward and inverse stochastic problems
• Applications of Unsupervised Deep Transfer Learning to Intelligent Fault Diagnosis: A Survey and Comparative Study
• Deep Exploration via Bootstrapped DQN
• The Challenge of Machine Learning in Space Weather Nowcasting and Forecasting
• Deep and Confident Prediction for Time Series at Uber
• Accurate Uncertainties for Deep Learning Using Calibrated Regression
• Modeling the Dynamics of PDE Systems with Physics-Constrained Deep Auto-Regressive Networks
• MILD-Net: Minimal Information Loss Dilated Network for Gland Instance Segmentation in Colon Histology Images
• Uncertainty-aware multi-view co-training for semi-supervised medical image segmentation and domain adaptation
• Neuromorphic Deep Learning Machines
• Unveiling the potential of Graph Neural Networks for network modeling and optimization in SDN
• Reliable deep-learning-based phase imaging with uncertainty quantification
• Bayesian Dark Knowledge
• Low-Rank Pairwise Alignment Bilinear Network For Few-Shot Fine-Grained Image Classification
• Human Activity Recognition using Recurrent Neural Networks
• SuperNNova: an open-source framework for Bayesian, Neural Network based supernova classification
• Uncertainty and Interpretability in Convolutional Neural Networks for Semantic Segmentation of Colorectal Polyps
• Learning Constitutive Relations from Indirect Observations Using Deep Neural Networks
• Exploration in Deep Reinforcement Learning: From Single-Agent to Multiagent Domain
• Towards Robust Evaluations of Continual Learning
• Prediction in ungauged regions with sparse flow duration curves and input-selection ensemble modeling
• Accelerating COVID-19 Differential Diagnosis with Explainable Ultrasound Image Analysis
• Automatic Brain Tumor Segmentation using Convolutional Neural Networks with Test-Time Augmentation
• Preconditioned Stochastic Gradient Langevin Dynamics for Deep Neural Networks
• Bayesian-Deep-Learning Estimation of Earthquake Location from Single-Station Observations
• High-Quality Prediction Intervals for Deep Learning: A Distribution-Free, Ensembled Approach
• The Fishyscapes Benchmark: Measuring Blind Spots in Semantic Segmentation
• Selective Synthetic Augmentation with HistoGAN for Improved Histopathology Image Classification
• Forecasting remaining useful life: Interpretable deep learning approach via variational Bayesian inferences
• Fusion of Probability Density Functions
• Indoor Scene Understanding in 2.5/3D for Autonomous Agents: A Survey
• Augmenting the Pathology Lab: An Intelligent Whole Slide Image Classification System for the Real World
• Deep convolutional neural networks for segmenting 3D in vivo multiphoton images of vasculature in Alzheimer disease mouse models
• Diminishing Uncertainty within the Training Pool: Active Learning for Medical Image Segmentation
• Deep Confidence: A Computationally Efficient Framework for Calculating Reliable Errors for Deep Neural Networks
• Copolymer Informatics with Multi-Task Deep Neural Networks
• Multivariate Confidence Calibration for Object Detection
• Continental-scale streamflow modeling of basins with reservoirs: towards a coherent deep-learning-based strategy
• Reset-free Trial-and-Error Learning for Robot Damage Recovery
• Adaptive Inference through Early-Exit Networks: Design, Challenges and Directions
• A deep learning approach to cosmological dark energy models
• SalsaNext: Fast, Uncertainty-aware Semantic Segmentation of LiDAR Point Clouds for Autonomous Driving
• Active learning of deep surrogates for PDEs: Application to metasurface design
• Implicit Full Waveform Inversion with Deep Neural Representation
• Probabilistic Forecasting of Sensory Data with Generative Adversarial Networks - ForGAN
• Bayesian deep learning for error estimation in the analysis of anomalous diffusion
• Bayesian Learning-Based Adaptive Control for Safety Critical Systems
• A Bayesian Deep Learning Framework for End-To-End Prediction of Emotion from Heartbeat
• 2022 Review of Data-Driven Plasma Science
• A machine learning approach for efficient uncertainty quantification using multiscale methods
• ViscNet: Neural network for predicting the fragility index and the temperature-dependency of viscosity
• Semi-Supervised Segmentation of Radiation-Induced Pulmonary Fibrosis from Lung CT Scans with Multi-Scale Guided Dense Attention
• Designing Silicon Photonic Devices using Artificial Neural Networks
• PsyPhy: A Psychophysics Driven Evaluation Framework for Visual Recognition
• Survey on Machine Learning for Traffic-Driven Service Provisioning in Optical Networks
• MCUa: Multi-level Context and Uncertainty aware Dynamic Deep Ensemble for Breast Cancer Histology Image Classification
• Hybrid neural network potential for multilayer graphene
• Fast Uncertainty Estimates in Deep Learning Interatomic Potentials
• Towards a Framework to Manage Perceptual Uncertainty for Safe Automated Driving
• Deep Learning Based Cloud Cover Parameterization for ICON
• Advanced Dropout: A Model-free Methodology for Bayesian Dropout Optimization
• Run-Time Monitoring of Machine Learning for Robotic Perception: A Survey of Emerging Trends
• Variational Auto-encoded Deep Gaussian Processes
• Unsupervised Domain Adversarial Self-Calibration for Electromyographic-based Gesture Recognition
• Combining distribution-based neural networks to predict weather forecast probabilities
• Reliable Prediction Errors for Deep Neural Networks Using Test-Time Dropout
• Robust Grasp Planning Over Uncertain Shape Completions
• An Artificial Intelligence Framework for Bidding Optimization with Uncertainty in Multiple Frequency Reserve Markets
• QU-BraTS: MICCAI BraTS 2020 Challenge on Quantifying Uncertainty in Brain Tumor Segmentation - Analysis of Ranking Scores and Benchmarking Results
• DLBI: Deep learning guided Bayesian inference for structure reconstruction of super-resolution fluorescence microscopy
• Deep Ensembles vs. Committees for Uncertainty Estimation in Neural-Network Force Fields: Comparison and Application to Active Learning
• CNN-Based Deep Learning in Solar Wind Forecasting
• Nowcasting-Nets: Deep Neural Network Structures for Precipitation Nowcasting Using IMERG
• Inspect, Understand, Overcome: A Survey of Practical Methods for AI Safety
• Precise Proximal Femur Fracture Classification for Interactive Training and Surgical Planning
• The Free Energy Principle for Perception and Action: A Deep Learning Perspective
• Ultrasound Signal Processing: From Models to Deep Learning
• Weakly Supervised Vessel Segmentation in X-ray Angiograms by Self-Paced Learning from Noisy Labels with Suggestive Annotation
• Deep Learning Convolutional Networks for Multiphoton Microscopy Vasculature Segmentation
• Detection of Gravitational Waves Using Bayesian Neural Networks
• Compensating for visibility artefacts in photoacoustic imaging with a deep learning approach providing prediction uncertainties
• Scalable Uncertainty Quantification for Deep Operator Networks using Randomized Priors
• Predictive Monitoring with Logic-Calibrated Uncertainty for Cyber-Physical Systems
• Fanaroff-Riley classification of radio galaxies using group-equivariant convolutional neural networks
• Artificial intelligence approaches for materials-by-design of energetic materials: state-of-the-art, challenges, and future directions
• CLeaR: An Adaptive Continual Learning Framework for Regression Tasks
• Attention U-Net as a surrogate model for groundwater prediction
• Mapping oil palm density at country scale: An active learning approach
• Uncertainty Averse Pushing with Model Predictive Path Integral Control
• Self-Paced Learning for Neural Machine Translation
• Unified Probabilistic Deep Continual Learning through Generative Replay and Open Set Recognition
• Learning Uncertainty with Artificial Neural Networks for Improved Predictive Process Monitoring
• PID-GAN: A GAN Framework based on a Physics-informed Discriminator for Uncertainty Quantification with Physics
• Molecule Identification with Rotational Spectroscopy and Probabilistic Deep Learning
• Human Action Performance using Deep Neuro-Fuzzy Recurrent Attention Model
• Calibration of Deep Probabilistic Models with Decoupled Bayesian Neural Networks
• Deep Gamblers: Learning to Abstain with Portfolio Theory
• Geometric Uncertainty in Patient-Specific Cardiovascular Modeling with Convolutional Dropout Networks
• Adaptive Prior Selection for Repertoire-based Online Adaptation in Robotics
• The Unreasonable Effectiveness of Deep Evidential Regression
• Uncertainty-Guided Efficient Interactive Refinement of Fetal Brain Segmentation from Stacks of MRI Slices
• Safe Control Synthesis with Uncertain Dynamics and Constraints
• Active Learning for Segmentation by Optimizing Content Information for Maximal Entropy
• Deep Distributional Time Series Models and the Probabilistic Forecasting of Intraday Electricity Prices
• Cognitive simulation models for inertial confinement fusion: Combining simulation and experimental data
• Toward Robust Image Classification
• CHS-Net: A Deep learning approach for hierarchical segmentation of COVID-19 infected CT images
• Machine Learning in High Energy Physics Community White Paper
• Short-term daily precipitation forecasting with seasonally-integrated autoencoder
• Information Aware Max-Norm Dirichlet Networks for Predictive Uncertainty Estimation
• Robust Simulation-Based Inference in Cosmology with Bayesian Neural Networks
• Parameterized Reinforcement Learning for Optical System Optimization
• Inseq: An Interpretability Toolkit for Sequence Generation Models
• Time-Dynamic Estimates of the Reliability of Deep Semantic Segmentation Networks
• Model-Based Policy Search Using Monte Carlo Gradient Estimation with Real Systems Application
• Into the Unknown: Active Monitoring of Neural Networks
• Reducing Drift in Visual Odometry by Inferring Sun Direction Using a Bayesian Convolutional Neural Network
• Learning Robotic Navigation from Experience: Principles, Methods, and Recent Results
• Towards Safe Machine Learning for CPS: Infer Uncertainty from Training Data
• Outside the Box: Abstraction-Based Monitoring of Neural Networks
• An Anomaly Detection Method for Satellites Using Monte Carlo Dropout
• Deep Posterior Distribution-based Embedding for Hyperspectral Image Super-resolution
• Bayesian Autoencoders for Drift Detection in Industrial Environments
• Concepts and Applications of Conformal Prediction in Computational Drug Discovery
• Mathematical Models of Overparameterized Neural Networks
• DropNeuron: Simplifying the Structure of Deep Neural Networks
• Operational Calibration: Debugging Confidence Errors for DNNs in the Field
• Active Visuo-Haptic Object Shape Completion
• NNVA: Neural Network Assisted Visual Analysis of Yeast Cell Polarization Simulation
• Efficient Bayesian Uncertainty Estimation for nnU-Net
• AANet: Attribute Attention Network for Person Re-Identifications
• Probabilistic Deep Learning to Quantify Uncertainty in Air Quality Forecasting
• Cleaning our own Dust: Simulating and Separating Galactic Dust Foregrounds with Neural Networks
• Deep Learning Analysis of Deeply Virtual Exclusive Photoproduction
• CertainNet: Sampling-free Uncertainty Estimation for Object Detection
• Twin Neural Network Regression
• Uncertainty-Guided Mixup for Semi-Supervised Domain Adaptation without Source Data
• Learning Scalable Deep Kernels with Recurrent Structure
• Physics-informed Bayesian inference of external potentials in classical density-functional theory
• CODEBench: A Neural Architecture and Hardware Accelerator Co-Design Framework
• Comparing Machine Learning and Interpolation Methods for Loop-Level Calculations
• STUaNet: Understanding uncertainty in spatiotemporal collective human mobility
• Prior and Posterior Networks: A Survey on Evidential Deep Learning Methods For Uncertainty Estimation
• Deeper Connections between Neural Networks and Gaussian Processes Speed-up Active Learning
• Bayesian Active Meta-Learning for Few Pilot Demodulation and Equalization
• A Self Supervised StyleGAN for Image Annotation and Classification with Extremely Limited Labels
• Self-Supervised Exploration via Disagreement
• System Theoretic View on Uncertainties
• Can We Leverage Predictive Uncertainty to Detect Dataset Shift and Adversarial Examples in Android Malware Detection?
• How Much Can We Really Trust You? Towards Simple, Interpretable Trust Quantification Metrics for Deep Neural Networks
• deepSIP: Linking Type Ia Supernova Spectra to Photometric Quantities with Deep Learning
• STAP: Sequencing Task-Agnostic Policies
• Confidence from Invariance to Image Transformations
• Quantifying Uncertainties in Natural Language Processing Tasks
• Dropout Distillation for Efficiently Estimating Model Confidence
• Unsupervised Knowledge-Transfer for Learned Image Reconstruction
• Cosmological Parameter Estimation and Inference using Deep Summaries
• Adversarial Attack for Uncertainty Estimation: Identifying Critical Regions in Neural Networks
• Target-mass Grasping of Entangled Food using Pre-grasping & Post-grasping
• Machine Learning and Cosmology
• Deep Network Uncertainty Maps for Indoor Navigation
• Brain Tumor Segmentation using 3D-CNNs with Uncertainty Estimation
• A Quantitative Comparison of Epistemic Uncertainty Maps Applied to Multi-Class Segmentation
• Bayesian Inference with Generative Adversarial Network Priors
• Bayesian Sparsification of Recurrent Neural Networks
• Stealing and Evading Malware Classifiers and Antivirus at Low False Positive Conditions
• Deep learning for Gaussian process tomography model selection using the ASDEX Upgrade SXR system
• Multi-organ segmentation: a progressive exploration of learning paradigms under scarce annotation
• LocalDrop: A Hybrid Regularization for Deep Neural Networks
• Deep Active Learning for Text Classification with Diverse Interpretations
• JuryGCN: Quantifying Jackknife Uncertainty on Graph Convolutional Networks
• Active Learning of Abstract Plan Feasibility
• Observation Space Matters: Benchmark and Optimization Algorithm
• Expectation Propagation for Poisson Data
• Treatment-aware Diffusion Probabilistic Model for Longitudinal MRI Generation and Diffuse Glioma Growth Prediction
• Learning Uncertainty For Safety-Oriented Semantic Segmentation In Autonomous Driving
• Gradient and Uncertainty Enhanced Sequential Sampling for Global Fit
• Deep cross-modality (MR-CT) educed distillation learning for cone beam CT lung tumor segmentation
• Pushing the bounds of dropout
• Double-Uncertainty Weighted Method for Semi-supervised Learning
• DS-UI: Dual-Supervised Mixture of Gaussian Mixture Models for Uncertainty Inference
• A deep learning-based framework for segmenting invisible clinical target volumes with estimated uncertainties for post-operative prostate cancer radiotherapy
• Quantifying Deep Learning Model Uncertainty in Conformal Prediction
• HYPPO: A Surrogate-Based Multi-Level Parallelism Tool for Hyperparameter Optimization
• Bayesian Neural Networks for Reversible Steganography
• Transfer learning driven design optimization for inertial confinement fusion
• Example Forgetting: A Novel Approach to Explain and Interpret Deep Neural Networks in Seismic Interpretation
• Improving Video Instance Segmentation by Light-weight Temporal Uncertainty Estimates
• Robust Node Classification on Graphs: Jointly from Bayesian Label Transition and Topology-based Label Propagation
• Edge-Assisted ML-Aided Uncertainty-Aware Vehicle Collision Avoidance at Urban Intersections
• Uncertainty Quantification in CNN Through the Bootstrap of Convex Neural Networks
• Estimating the Uncertainty in Emotion Attributes using Deep Evidential Regression
• Scan-specific Self-supervised Bayesian Deep Non-linear Inversion for Undersampled MRI Reconstruction
• Deep Bayesian ICP Covariance Estimation
• Towards Reducing Aleatoric Uncertainty for Medical Imaging Tasks
• Quantifying sources of uncertainty in drug discovery predictions with probabilistic models
• Exploring Bayesian Deep Learning for Urgent Instructor Intervention Need in MOOC Forums
• Dependency-Aware Named Entity Recognition with Relative and Global Attentions
• Detecting Adversarial Examples in Convolutional Neural Networks
• Bayesian Dropout
• Integrating Uncertainty into Neural Network-based Speech Enhancement
• Confidence-Calibrated Face and Kinship Verification
• Medical Coding with Biomedical Transformer Ensembles and Zero/Few-shot Learning
• Accelerating Monte Carlo Bayesian Inference via Approximating Predictive Uncertainty over Simplex
• Data Uncertainty Learning in Face Recognition
• Augmenting semantic lexicons using word embeddings and transfer learning
• Long-Term Visitation Value for Deep Exploration in Sparse Reward Reinforcement Learning
• AI-Driven Discovery of High Performance Polymer Electrodes for Next-Generation Batteries
• Uncertainty-Aware Self-supervised Neural Network for Liver $T_{1\rho }$ Mapping with Relaxation Constraint
• Uncertainty-Encoded Multi-Modal Fusion for Robust Object Detection in Autonomous Driving
• Universal uncertainty estimation for nuclear detector signals with neural networks and ensemble learning
• Interpretable Uncertainty Quantification in AI for HEP
• Influence of uncertainty estimation techniques on false-positive reduction in liver lesion detection
• Shift-BNN: Highly-Efficient Probabilistic Bayesian Neural Network Training via Memory-Friendly Pattern Retrieving
• Multi-Objective Optimization Under Uncertainty of Part Quality in Fused Filament Fabrication
• On the use of uncertainty in classifying Aedes Albopictus mosquitoes
• Dataset Complexity Assessment Based on Cumulative Maximum Scaled Area Under Laplacian Spectrum
• Closer Look at the Uncertainty Estimation in Semantic Segmentation under Distributional Shift
• Learning Probabilistic Coordinate Fields for Robust Correspondences
• Adversarial Artifact Detection in EEG-Based Brain-Computer Interfaces
• Tire-road friction estimation and uncertainty assessment to improve electric aircraft braking system
• A Dataset-free Deep learning Method for Low-Dose CT Image Reconstruction
• Interpretation of Deep Temporal Representations by Selective Visualization of Internally Activated Nodes
• Scalable Natural Gradient Langevin Dynamics in Practice
• Sparseout: Controlling Sparsity in Deep Networks
• Gaussian Gated Linear Networks
• Bayesian Neural Networks at Scale: A Performance Analysis and Pruning Study
• Ranking over Regression for Bayesian Optimization and Molecule Selection
• A Closed-Form Uncertainty Propagation in Non-Rigid Structure from Motion
• Theoretical characterization of uncertainty in high-dimensional linear classification
• Delving Deeper into the Decoder for Video Captioning
• Vision-Based Uncertainty-Aware Motion Planning based on Probabilistic Semantic Segmentation
• A Trustworthiness Score to Evaluate DNN Predictions
• An Effective, Robust and Fairness-aware Hate Speech Detection Framework
• Towards Balanced Active Learning for Multimodal Classification
• Better with Less: A Data-Active Perspective on Pre-Training Graph Neural Networks
• Towards Expressive Priors for Bayesian Neural Networks: Poisson Process Radial Basis Function Networks
• Confidence-Guided Learning Process for Continuous Classification of Time Series
• Model Monitoring and Dynamic Model Selection in Travel Time-series Forecasting
• Bayesian Graph Neural Network for Fast identification of critical nodes in Uncertain Complex Networks
• Deep Anti-Regularized Ensembles provide reliable out-of-distribution uncertainty quantification
• Quantification of Uncertainties in Deep Learning-based Environment Perception
• A t-distribution based operator for enhancing out of distribution robustness of neural network classifiers
• Bayesian posterior approximation with stochastic ensembles
• Probabilistically-autoencoded horseshoe-disentangled multidomain item-response theory models
• AutoDEUQ: Automated Deep Ensemble with Uncertainty Quantification
• Inferring Javascript types using Graph Neural Networks
• Bayesian Neural Network Ensembles
• Probabilistic Spatial Transformer Networks
• A machine learning photon detection algorithm for coherent X-ray ultrafast fluctuation analysis
• Effect of latent space distribution on the segmentation of images with multiple annotations
• Insights into Fairness through Trust: Multi-scale Trust Quantification for Financial Deep Learning
• A Bayesian neural network predicts the dissolution of compact planetary systems
• Differentially Private Dropout
• Towards a Kernel based Uncertainty Decomposition Framework for Data and Models
• False Negative Reduction in Video Instance Segmentation using Uncertainty Estimates
• Evolving Neural Selection with Adaptive Regularization
• Variational Smoothing in Recurrent Neural Network Language Models
• On the Robustness of Monte Carlo Dropout Trained with Noisy Labels
• The Monte Carlo Transformer: a stochastic self-attention model for sequence prediction
• Nonparametric Bayesian Deep Networks with Local Competition
• Bayesian Transformer Language Models for Speech Recognition
• Give me (un)certainty -- An exploration of parameters that affect segmentation uncertainty
• Active Learning under Label Shift
• Hidden Markov Neural Networks
• Information-Theoretic Odometry Learning
• Guiding the retraining of convolutional neural networks against adversarial inputs
• Label uncertainty-guided multi-stream model for disease screening
• Using Artificial Populations to Study Psychological Phenomena in Neural Models
• Confident Naturalness Explanation (CNE): A Framework to Explain and Assess Patterns Forming Naturalness
• Leveraging Active Subspaces to Capture Epistemic Model Uncertainty in Deep Generative Models for Molecular Design
• Improving Uncertainty-Error Correspondence in Deep Bayesian Medical Image Segmentation
• Geometrically Enriched Latent Spaces
• A Comparison of Mesh-Free Differentiable Programming and Data-Driven Strategies for Optimal Control under PDE Constraints
• Imitation Learning of Robot Policies by Combining Language, Vision and Demonstration
• Fully Bayesian Recurrent Neural Networks for Safe Reinforcement Learning
• Black-Box Data-efficient Policy Search for Robotics
• Surprising properties of dropout in deep networks
• Deep kernel processes
• Entropy-based Active Learning of Graph Neural Network Surrogate Models for Materials Properties
• Detecting Out-of-distribution Objects Using Neuron Activation Patterns
• Online Black-Box Confidence Estimation of Deep Neural Networks
• Deep Momentum Uncertainty Hashing
• 3D extinction mapping of the Milky Way using Convolutional Neural Networks: Presentation of the method and demonstration in the Carina Arm region
• That Label's Got Style: Handling Label Style Bias for Uncertain Image Segmentation
• Uncertainty Propagation in Node Classification
• Performance Measurement for Deep Bayesian Neural Network
• Improved Image Matting via Real-time User Clicks and Uncertainty Estimation
• The role of MRI physics in brain segmentation CNNs: achieving acquisition invariance and instructive uncertainties
• Improving compute efficacy frontiers with SliceOut
• A Bayesian Convolutional Neural Network for Robust Galaxy Ellipticity Regression
• Deep Out-of-Distribution Uncertainty Quantification via Weight Entropy Maximization
• Quantification of Predictive Uncertainty via Inference-Time Sampling
• Estimation of Counterfactual Interventions under Uncertainties
• Bayesian Uncertainty and Expected Gradient Length -- Regression: Two Sides Of The Same Coin?
• Dual-Teacher++: Exploiting Intra-domain and Inter-domain Knowledge with Reliable Transfer for Cardiac Segmentation
• Localized Uncertainty Attacks
• Confidence Contours: Uncertainty-Aware Annotation for Medical Semantic Segmentation
• REVE: Regularizing Deep Learning with Variational Entropy Bound
• Improve Uncertainty Estimation for Unknown Classes in Bayesian Neural Networks with Semi-Supervised /One Set Classification
• The Robust Semantic Segmentation UNCV2023 Challenge Results
• Law of Large Numbers for Bayesian two-layer Neural Network trained with Variational Inference
• Variational Bayesian Sequence-to-Sequence Networks for Memory-Efficient Sign Language Translation
• Bayesian Metric Learning for Uncertainty Quantification in Image Retrieval
• Selective Probabilistic Classifier Based on Hypothesis Testing
• "None of the Above":Measure Uncertainty in Dialog Response Retrieval
• Techniques Toward Optimizing Viewability in RTB Ad Campaigns Using Reinforcement Learning
• Langevin Monte Carlo for strongly log-concave distributions: Randomized midpoint revisited
• Architectural Resilience to Foreground-and-Background Adversarial Noise
• MACEst: The reliable and trustworthy Model Agnostic Confidence Estimator
• On the generalization of bayesian deep nets for multi-class classification
• Riemannian Laplace approximations for Bayesian neural networks