Decomposition of Uncertainty in Bayesian Deep Learning for Efficient and Risk-sensitive Learning

Abstract

Bayesian neural networks with latent variables are scalable and flexible probabilistic models: They account for uncertainty in the estimation of the network weights and, by making use of latent variables, can capture complex noise patterns in the data. We show how to extract and decompose uncertainty into epistemic and aleatoric components for decision-making purposes. This allows us to successfully identify informative points for active learning of functions with heteroscedastic and bimodal noise. Using the decomposition we further define a novel risk-sensitive criterion for reinforcement learning to identify policies that balance expected cost, model-bias and noise aversion.

arXiv commentThis paper supersedes arXiv:1706.08495

Cited by in corpus (45)

• Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methods
• Hands-on Bayesian Neural Networks -- a Tutorial for Deep Learning Users
• A Review and Comparative Study on Probabilistic Object Detection in Autonomous Driving
• Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models
• Model-Predictive Policy Learning with Uncertainty Regularization for Driving in Dense Traffic
• Uncertainty Aware Semi-Supervised Learning on Graph Data
• Estimating Risk and Uncertainty in Deep Reinforcement Learning
• Getting a CLUE: A Method for Explaining Uncertainty Estimates
• On Hallucination and Predictive Uncertainty in Conditional Language Generation
• Benchmarking Bayesian Deep Learning on Diabetic Retinopathy Detection Tasks
• Model-based Reinforcement Learning from Signal Temporal Logic Specifications
• Predictive Uncertainty Quantification with Compound Density Networks
• Dense Uncertainty Estimation
• Graph Posterior Network: Bayesian Predictive Uncertainty for Node Classification
• Influence of uncertainty estimation techniques on false-positive reduction in liver lesion detection
• Quantifying Sources of Uncertainty in Deep Learning-Based Image Reconstruction
• Epistemic Risk-Sensitive Reinforcement Learning
• Fast and reliable uncertainty quantification with neural network ensembles for industrial image classification
• Deep Conditional Transformation Models
• Improving Robustness of Deep Neural Networks for Aerial Navigation by Incorporating Input Uncertainty
• Bayesian OOD detection with aleatoric uncertainty and outlier exposure
• Posterior Meta-Replay for Continual Learning
• Robust exploration in linear quadratic reinforcement learning
• Exploring Covariate and Concept Shift for Detection and Calibration of Out-of-Distribution Data
• SENTINEL: Taming Uncertainty with Ensemble-based Distributional Reinforcement Learning
• Mitigating the Effects of Non-Identifiability on Inference for Bayesian Neural Networks with Latent Variables
• Approaching Neural Network Uncertainty Realism
• Uncertainty Estimation and Calibration with Finite-State Probabilistic RNNs
• Iterative Distillation for Better Uncertainty Estimates in Multitask Emotion Recognition
• A Bayesian Convolutional Neural Network for Robust Galaxy Ellipticity Regression
• Learning to Plan Optimistically: Uncertainty-Guided Deep Exploration via Latent Model Ensembles
• Style-based Composer Identification and Attribution of Symbolic Music Scores: a Systematic Survey
• Model-based micro-data reinforcement learning: what are the crucial model properties and which model to choose?
• A Tale Of Two Long Tails
• Collaborative Uncertainty in Multi-Agent Trajectory Forecasting
• Bayesian Uncertainty and Expected Gradient Length -- Regression: Two Sides Of The Same Coin?
• Risk Sensitive Model-Based Reinforcement Learning using Uncertainty Guided Planning
• A Bayesian Neural Network based on Dropout Regulation
• Learning to Predict Error for MRI Reconstruction
• Greedy Bayesian Posterior Approximation with Deep Ensembles
• Deep Classifiers with Label Noise Modeling and Distance Awareness
• A General Divergence Modeling Strategy for Salient Object Detection
• Dense Uncertainty Estimation via an Ensemble-based Conditional Latent Variable Model
• Boosting Cross-Lingual Transfer via Self-Learning with Uncertainty Estimation
• Self-Paced Uncertainty Estimation for One-shot Person Re-Identification