A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges (2020-11-12T00:00:00.000000Z)

TL;DR

This study reviews recent advances in UQ methods used in deep learning and investigates the application of these methods in reinforcement learning (RL), and outlines a few important applications of UZ methods.

Authors

Farhad Pourpanah

2 papers

Moloud Abdar

1 papers

Sadiq Hussain

1 papers

Dana Rezazadegan

1 papers

Li Liu

1 papers

M. Ghavamzadeh

1 papers

P. Fieguth

5 papers

Xiaochun Cao

3 papers

A. Khosravi

2 papers

U. R. Acharya

1 papers

V. Makarenkov

1 papers

S. Nahavandi

3 papers

TL;DR

Authors

References579 items

Harnessing Uncertainty in Domain Adaptation for MRI Prostate Lesion Segmentation

An Effective Data Refinement Approach for Upper Gastrointestinal Anatomy Recognition

Efficient Shapley Explanation for Features Importance Estimation Under Uncertainty

Double-Uncertainty Weighted Method for Semi-supervised Learning

Uncertainty Estimates as Data Selection Criteria to Boost Omni-Supervised Learning

Deep Reinforcement Active Learning for Medical Image Classification

Cartilage Segmentation in High-Resolution 3D Micro-CT Images via Uncertainty-Guided Self-training with Very Sparse Annotation

Heterogeneity Measurement of Cardiac Tissues Leveraging Uncertainty Information from Image Segmentation

RevPHiSeg: A Memory-Efficient Neural Network for Uncertainty Quantification in Medical Image Segmentation

Mixup-CAM: Weakly-supervised Semantic Segmentation via Uncertainty Regularization

Difficulty-aware Glaucoma Classification with Multi-Rater Consensus Modeling

A Bayesian Hierarchical Network for Combining Heterogeneous Data Sources in Medical Diagnoses

Self-Loop Uncertainty: A Novel Pseudo-Label for Semi-Supervised Medical Image Segmentation

Integrating uncertainty in deep neural networks for MRI based stroke analysis

Prediction Intervals: Split Normal Mixture from Quality-Driven Deep Ensembles

Transferable Calibration with Lower Bias and Variance in Domain Adaptation

Hands-On Bayesian Neural Networks—A Tutorial for Deep Learning Users

On uncertainty estimation in active learning for image segmentation

SDE-Net: Equipping Deep Neural Networks with Uncertainty Estimates

Exploiting Uncertainties from Ensemble Learners to Improve Decision-Making in Healthcare AI

Bayesian Deep Ensembles via the Neural Tangent Kernel

Revisiting One-vs-All Classifiers for Predictive Uncertainty and Out-of-Distribution Detection in Neural Networks

Few Is Enough: Task-Augmented Active Meta-Learning for Brain Cell Classification

Uncertainty Quantification in Deep Residual Neural Networks

SpinalNet: Deep Neural Network With Gradual Input

Single Shot MC Dropout Approximation

Few-shot Relation Extraction via Bayesian Meta-learning on Relation Graphs

Evaluating Uncertainty Estimation Methods on 3D Semantic Segmentation of Point Clouds

Meta-Learning Stationary Stochastic Process Prediction with Convolutional Neural Processes

Ranking Information Extracted from Uncertainty Quantification of the Prediction of a Deep Learning Model on Medical Time Series Data

Distributed Selection of Continuous Features in Multilabel Classification Using Mutual Information

Identifying Causal Effect Inference Failure with Uncertainty-Aware Models

Discriminative Jackknife: Quantifying Uncertainty in Deep Learning via Higher-Order Influence Functions

Uncertainty-aware multi-view co-training for semi-supervised medical image segmentation and domain adaptation

Can Autonomous Vehicles Identify, Recover From, and Adapt to Distribution Shifts?

Probabilistic inference of Bayesian neural networks with generalized expectation propagation

Deep Q-Network-Driven Catheter Segmentation in 3D US by Hybrid Constrained Semi-Supervised Learning and Dual-UNet

Clinical Risk Prediction with Temporal Probabilistic Asymmetric Multi-Task Learning

Stacking for Non-mixing Bayesian Computations: The Curse and Blessing of Multimodal Posteriors

MUMBO: MUlti-task Max-value Bayesian Optimization

Learned Uncertainty-Aware (LUNA) Bases for Bayesian Regression using Multi-Headed Auxiliary Networks

A Bayesian Evaluation Framework for Ground Truth-Free Visual Recognition Tasks

Calibration of Model Uncertainty for Dropout Variational Inference

Variational Bayesian Monte Carlo with Noisy Likelihoods

Neural Ensemble Search for Performant and Calibrated Predictions

NADS: Neural Architecture Distribution Search for Uncertainty Awareness

Multi-View Representation Learning With Deep Gaussian Processes

Uncertainty-aware domain alignment for anatomical structure segmentation

Wat zei je? Detecting Out-of-Distribution Translations with Variational Transformers

Bayesian Graph Neural Networks with Adaptive Connection Sampling

Uncertainty quantification in medical image segmentation with Normalizing Flows

Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awareness

Uncertainty Estimation in Deep Neural Networks for Dermoscopic Image Classification

Uncertainty-Aware CNNs for Depth Completion: Uncertainty from Beginning to End

Generalizing Hand Segmentation in Egocentric Videos With Uncertainty-Guided Model Adaptation

Calibrated Reliable Regression using Maximum Mean Discrepancy

Stochastic Segmentation Networks: Modelling Spatially Correlated Aleatoric Uncertainty

Automated Detection of Presymptomatic Conditions in Spinocerebellar Ataxia Type 2 Using Monte Carlo Dropout and Deep Neural Network Techniques with Electrooculogram Signals

Improving Regression Uncertainty Estimates with an Empirical Prior

CRUDE: Calibrating Regression Uncertainty Distributions Empirically

Learning To Classify Images Without Labels

Pairwise Supervised Hashing with Bernoulli Variational Auto-Encoder and Self-Control Gradient Estimator

Efficient Ensemble Model Generation for Uncertainty Estimation with Bayesian Approximation in Segmentation

Uncertainty Estimation in Deep 2D Echocardiography Segmentation

Global inducing point variational posteriors for Bayesian neural networks and deep Gaussian processes

Efficient and Scalable Bayesian Neural Nets with Rank-1 Factors

On the Uncertainty of Self-Supervised Monocular Depth Estimation

Simultaneous Perturbation Stochastic Approximation for Few-Shot Learning

Empirical Bayes Transductive Meta-Learning with Synthetic Gradients

Amortized Bayesian Model Comparison With Evidential Deep Learning

Uncertainty-Aware Consistency Regularization for Cross-Domain Semantic Segmentation

Deeply uncertain: comparing methods of uncertainty quantification in deep learning algorithms

UC-Net: Uncertainty Inspired RGB-D Saliency Detection via Conditional Variational Autoencoders

Regularizing Meta-Learning via Gradient Dropout

Uncertainty-Aware Deep Classifiers Using Generative Models

Uncertainty Aware Graph Gaussian Process for Semi-Supervised Learning

Uncertainty-Aware Search Framework for Multi-Objective Bayesian Optimization