Looking at the posterior: on the origin of uncertainty in neural-network classification

11/26/2022
by   H. Linander, et al.
0

Bayesian inference can quantify uncertainty in the predictions of neural networks using posterior distributions for model parameters and network output. By looking at these posterior distributions, one can separate the origin of uncertainty into aleatoric and epistemic. We use the joint distribution of predictive uncertainty and epistemic uncertainty to quantify how this interpretation of uncertainty depends upon model architecture, dataset complexity, and data distributional shifts in image classification tasks. We conclude that the origin of uncertainty is subjective to each neural network and that the quantification of the induced uncertainty from data distributional shifts depends on the complexity of the underlying dataset. Furthermore, we show that the joint distribution of predictive and epistemic uncertainty can be used to identify data domains where the model is most accurate. To arrive at these results, we use two common posterior approximation methods, Monte-Carlo dropout and deep ensembles, for fully-connected, convolutional and attention-based neural networks.

READ FULL TEXT
research
10/26/2021

AutoDEUQ: Automated Deep Ensemble with Uncertainty Quantification

Deep neural networks are powerful predictors for a variety of tasks. How...
research
07/01/2021

On the Practicality of Deterministic Epistemic Uncertainty

A set of novel approaches for estimating epistemic uncertainty in deep n...
research
02/18/2023

Approximate Thompson Sampling via Epistemic Neural Networks

Thompson sampling (TS) is a popular heuristic for action selection, but ...
research
10/09/2021

Evaluating Predictive Distributions: Does Bayesian Deep Learning Work?

Posterior predictive distributions quantify uncertainties ignored by poi...
research
06/16/2020

Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Counts

Accurate estimation of aleatoric and epistemic uncertainty is crucial to...
research
01/07/2022

Bayesian Neural Networks for Reversible Steganography

Recent advances in deep learning have led to a paradigm shift in reversi...
research
11/27/2018

Bayesian Neural Network Ensembles

Ensembles of neural networks (NNs) have long been used to estimate predi...

Please sign up or login with your details

Forgot password? Click here to reset