DAUX: a Density-based Approach for Uncertainty eXplanations

07/11/2022
by   Hao Sun, et al.
18

Uncertainty quantification (UQ) is essential for creating trustworthy machine learning models. Recent years have seen a steep rise in UQ methods that can flag suspicious examples, however, it is often unclear what exactly these methods identify. In this work, we propose an assumption-light method for interpreting UQ models themselves. We introduce the confusion density matrix – a kernel-based approximation of the misclassification density – and use this to categorize suspicious examples identified by a given UQ method into three classes: out-of-distribution (OOD) examples, boundary (Bnd) examples, and examples in regions of high in-distribution misclassification (IDM). Through extensive experiments, we shed light on existing UQ methods and show that the cause of the uncertainty differs across models. Additionally, we show how the proposed framework can make use of the categorized examples to improve predictive performance.

READ FULL TEXT

page 2

page 5

page 7

page 12

research
10/18/2022

Uncertainty in Extreme Multi-label Classification

Uncertainty quantification is one of the most crucial tasks to obtain tr...
research
02/04/2019

Predictive Uncertainty Quantification with Compound Density Networks

Despite the huge success of deep neural networks (NNs), finding good mec...
research
10/05/2022

Explanation Uncertainty with Decision Boundary Awareness

Post-hoc explanation methods have become increasingly depended upon for ...
research
09/19/2023

Adversarial Attacks Against Uncertainty Quantification

Machine-learning models can be fooled by adversarial examples, i.e., car...
research
10/30/2022

A view on model misspecification in uncertainty quantification

Estimating uncertainty of machine learning models is essential to assess...
research
10/20/2020

Towards Maximizing the Representation Gap between In-Domain & Out-of-Distribution Examples

Among existing uncertainty estimation approaches, Dirichlet Prior Networ...
research
07/16/2021

An Uncertainty-Aware, Shareable and Transparent Neural Network Architecture for Brain-Age Modeling

The deviation between chronological age and age predicted from neuroimag...

Please sign up or login with your details

Forgot password? Click here to reset