{\delta}-CLUE: Diverse Sets of Explanations for Uncertainty Estimates

TL;DR

This paper introduces -CLUE, an extension of CLUE that generates diverse sets of plausible explanations within a -ball in latent space, enhancing interpretability of uncertainty estimates from probabilistic models.

Contribution

The paper proposes -CLUE, a method to produce multiple diverse explanations for model confidence, addressing the limitation of single explanations in previous CLUE approaches.

Findings

Generates diverse, plausible explanations within a -ball in latent space.

Enhances interpretability of uncertainty estimates from probabilistic models.

Provides a set of explanations rather than a single counterfactual.

Abstract

To interpret uncertainty estimates from differentiable probabilistic models, recent work has proposed generating Counterfactual Latent Uncertainty Explanations (CLUEs). However, for a single input, such approaches could output a variety of explanations due to the lack of constraints placed on the explanation. Here we augment the original CLUE approach, to provide what we call $\delta$-CLUE. CLUE indicates $\it{one}$ way to change an input, while remaining on the data manifold, such that the model becomes more confident about its prediction. We instead return a $\it{set}$ of plausible CLUEs: multiple, diverse inputs that are within a $\delta$ ball of the original input in latent space, all yielding confident predictions.