On the Calibration of Epistemic Uncertainty: Principles, Paradoxes and Conflictual Loss

• URL: http://arxiv.org/abs/2407.12211v1
• Date: Tue, 16 Jul 2024 23:21:28 GMT
• Title: On the Calibration of Epistemic Uncertainty: Principles, Paradoxes and Conflictual Loss
• Authors: Mohammed Fellaji, Frédéric Pennerath, Brieuc Conan-Guez, Miguel Couceiro,
• Abstract summary: Evidential uncertainty is produced by Deep Ensembles, Bayesian Deep Networks, or Evidential Deep Networks. Although measurable, this form of uncertainty is difficult to calibrate on an objective basis. We propose a regularization function for deep ensembles, called conflictual loss in line with the above requirements.
• Score: 3.8248583585487155
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The calibration of predictive distributions has been widely studied in deep learning, but the same cannot be said about the more specific epistemic uncertainty as produced by Deep Ensembles, Bayesian Deep Networks, or Evidential Deep Networks. Although measurable, this form of uncertainty is difficult to calibrate on an objective basis as it depends on the prior for which a variety of choices exist. Nevertheless, epistemic uncertainty must in all cases satisfy two formal requirements: first, it must decrease when the training dataset gets larger and, second, it must increase when the model expressiveness grows. Despite these expectations, our experimental study shows that on several reference datasets and models, measures of epistemic uncertainty violate these requirements, sometimes presenting trends completely opposite to those expected. These paradoxes between expectation and reality raise the question of the true utility of epistemic uncertainty as estimated by these models. A formal argument suggests that this disagreement is due to a poor approximation of the posterior distribution rather than to a flaw in the measure itself. Based on this observation, we propose a regularization function for deep ensembles, called conflictual loss in line with the above requirements. We emphasize its strengths by showing experimentally that it restores both requirements of epistemic uncertainty, without sacrificing either the performance or the calibration of the deep ensembles.

Related papers

• (Implicit) Ensembles of Ensembles: Epistemic Uncertainty Collapse in Large Models [3.0539022029583953]
  Epistemic uncertainty is crucial for safety-critical applications and out-of-distribution detection tasks. We uncover a paradoxical phenomenon in deep learning models: an epistemic uncertainty collapse as model complexity increases.
  arXiv  Detail & Related papers  (2024-09-04T11:45:55Z)
• The Epistemic Uncertainty Hole: an issue of Bayesian Neural Networks [0.6906005491572401]
  We show that the evolution of "epistemic uncertainty metrics" regarding the model size and the size of the training set, goes against theoretical expectations. This phenomenon, which we call "epistemic uncertainty hole", is all the more problematic as it undermines the entire applicative potential of BDL.
  arXiv  Detail & Related papers  (2024-07-02T06:54:46Z)
• Quantification of Uncertainty with Adversarial Models [6.772632213236167]
  Quantifying uncertainty is important for actionable predictions in real-world applications. We suggest Quantification of Uncertainty with Adversarial Models (QUAM) QUAM identifies regions where the whole product under the integral is large, not just the posterior.
  arXiv  Detail & Related papers  (2023-07-06T17:56:10Z)
• The Implicit Delta Method [61.36121543728134]
  In this paper, we propose an alternative, the implicit delta method, which works by infinitesimally regularizing the training loss of uncertainty. We show that the change in the evaluation due to regularization is consistent for the variance of the evaluation estimator, even when the infinitesimal change is approximated by a finite difference.
  arXiv  Detail & Related papers  (2022-11-11T19:34:17Z)
• Monotonicity and Double Descent in Uncertainty Estimation with Gaussian Processes [52.92110730286403]
  It is commonly believed that the marginal likelihood should be reminiscent of cross-validation metrics and that both should deteriorate with larger input dimensions. We prove that by tuning hyper parameters, the performance, as measured by the marginal likelihood, improves monotonically with the input dimension. We also prove that cross-validation metrics exhibit qualitatively different behavior that is characteristic of double descent.
  arXiv  Detail & Related papers  (2022-10-14T08:09:33Z)
• Uncertainty Quantification for Traffic Forecasting: A Unified Approach [21.556559649467328]
  Uncertainty is an essential consideration for time series forecasting tasks. In this work, we focus on quantifying the uncertainty of traffic forecasting. We develop Deep S-Temporal Uncertainty Quantification (STUQ), which can estimate both aleatoric and relational uncertainty.
  arXiv  Detail & Related papers  (2022-08-11T15:21:53Z)
• Dense Uncertainty Estimation via an Ensemble-based Conditional Latent Variable Model [68.34559610536614]
  We argue that the aleatoric uncertainty is an inherent attribute of the data and can only be correctly estimated with an unbiased oracle model. We propose a new sampling and selection strategy at train time to approximate the oracle model for aleatoric uncertainty estimation. Our results show that our solution achieves both accurate deterministic results and reliable uncertainty estimation.
  arXiv  Detail & Related papers  (2021-11-22T08:54:10Z)
• Dense Uncertainty Estimation [62.23555922631451]
  In this paper, we investigate neural networks and uncertainty estimation techniques to achieve both accurate deterministic prediction and reliable uncertainty estimation. We work on two types of uncertainty estimations solutions, namely ensemble based methods and generative model based methods, and explain their pros and cons while using them in fully/semi/weakly-supervised framework.
  arXiv  Detail & Related papers  (2021-10-13T01:23:48Z)
• The Hidden Uncertainty in a Neural Networks Activations [105.4223982696279]
  The distribution of a neural network's latent representations has been successfully used to detect out-of-distribution (OOD) data. This work investigates whether this distribution correlates with a model's epistemic uncertainty, thus indicating its ability to generalise to novel inputs.
  arXiv  Detail & Related papers  (2020-12-05T17:30:35Z)
• Discriminative Jackknife: Quantifying Uncertainty in Deep Learning via Higher-Order Influence Functions [121.10450359856242]
  We develop a frequentist procedure that utilizes influence functions of a model's loss functional to construct a jackknife (or leave-one-out) estimator of predictive confidence intervals. The DJ satisfies (1) and (2), is applicable to a wide range of deep learning models, is easy to implement, and can be applied in a post-hoc fashion without interfering with model training or compromising its accuracy.
  arXiv  Detail & Related papers  (2020-06-29T13:36:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.