Pith. sign in

REVIEW 6 cited by

Getting a CLUE: A Method for Explaining Uncertainty Estimates

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2006.06848 v2 pith:DQZCCDVL submitted 2020-06-11 stat.ML cs.LG

Getting a CLUE: A Method for Explaining Uncertainty Estimates

Javier Antor\'an , Umang Bhatt , Tameem Adel , Adrian Weller , Jos\'e Miguel Hern\'andez-Lobato This is my paper
classification stat.ML cs.LG
keywords uncertaintyclueinputmethodcounterfactualestimatesexperimentsexplanations
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Both uncertainty estimation and interpretability are important factors for trustworthy machine learning systems. However, there is little work at the intersection of these two areas. We address this gap by proposing a novel method for interpreting uncertainty estimates from differentiable probabilistic models, like Bayesian Neural Networks (BNNs). Our method, Counterfactual Latent Uncertainty Explanations (CLUE), indicates how to change an input, while keeping it on the data manifold, such that a BNN becomes more confident about the input's prediction. We validate CLUE through 1) a novel framework for evaluating counterfactual explanations of uncertainty, 2) a series of ablation experiments, and 3) a user study. Our experiments show that CLUE outperforms baselines and enables practitioners to better understand which input patterns are responsible for predictive uncertainty.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. An Explainable Gaussian Process Auto-encoder for Tabular Data

    cs.LG 2025-08 conditional novelty 6.0

    A Gaussian-process autoencoder with a latent-space density estimator generates counterfactual examples for tabular data, with competitive or better scores on several evaluation metrics.

  2. Optimized Instance Alteration for Explaining and Assessing Robustness of Classifiers

    cs.LG 2026-07 conditional novelty 5.0

    An explainability-aware L0 penalty yields coherent counterfactual edits, and the same geometry defines a Tolerance-Region Confusion Matrix that quantifies class-to-class fragility under interpretable perturbations.

  3. A Framework for Variational Inference of Lightweight Bayesian Neural Networks with Heteroscedastic Uncertainties

    cs.LG 2024-02 unverdicted novelty 5.0

    Framework embeds aleatoric and epistemic uncertainties into BNN parameter variances and applies moment propagation for sampling-free variational inference in lightweight networks.

  4. Target-confidence Recourse Using tSeTlin machines: TRUST

    cs.LG 2026-06 unverdicted novelty 4.0

    TRUST searches for minimal input changes that achieve a user-defined confidence target in PTM models, claiming perfect robustness and low cost on benchmarks versus standard boundary-crossing methods.

  5. Uncertainty Awareness and Trust in Explainable AI- On Trust Calibration using Local and Global Explanations

    cs.AI 2025-09 conditional novelty 4.0

    People who saw uncertainty visualizations trusted a more certain model more, but the study does not demonstrate that this explanation calibrates trust better than numeric accuracy.

  6. Tabular Diffusion Counterfactual Explanations

    cs.LG 2025-08 conditional novelty 4.0

    TDCE guides a tabular diffusion reverse process with Gumbel-softmax classifier gradients to produce counterfactual explanations, achieving high validity on four benchmarks.