DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Can a Single Neuron Learn Predictive Uncertainty?

Solano Carrillo, Edgardo (2023) Can a Single Neuron Learn Predictive Uncertainty? International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 31 (3), pp. 471-496. World Scientific. doi: 10.1142/S021848852350023X. ISSN 0218-4885.

[img] PDF - Preprint version (submitted draft)

Official URL: https://www.worldscientific.com/doi/epdf/10.1142/S021848852350023X


Uncertainty estimation methods using deep learning approaches strive against separating how uncertain the state of the world manifests to us via measurement (objective end) from the way this gets scrambled with the model specification and training procedure used to predict such state (subjective means) — e.g., number of neurons, depth, connections, priors (if the model is bayesian), weight initialization, etc. This poses the question of the extent to which one can eliminate the degrees of freedom associated with these specifications and still being able to capture the objective end. Here, a novel nonparametric quantile estimation method for continuous random variables is introduced, based on the simplest neural network architecture with one degree of freedom: a single neuron. Its advantage is first shown in synthetic experiments comparing with the quantile estimation achieved from ranking the order statistics (specifically for small sample size) and with quantile regression. In real-world applications, the method can be used to quantify predictive uncertainty under the split conformal prediction setting, whereby prediction intervals are estimated from the residuals of a pre-trained model on a heldout validation set and then used to quantify the uncertainty in future predictions — the single neuron used here as a structureless “thermometer” that measures how uncertain the pre-trained model is. Benchmarking regression and classification experiments demonstrate that the method is competitive in quality and coverage with state-of-the-art solutions, with the added benefit of being more computationally efficient.

Item URL in elib:https://elib.dlr.de/195849/
Document Type:Article
Title:Can a Single Neuron Learn Predictive Uncertainty?
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Date:June 2023
Journal or Publication Title:International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In ISI Web of Science:Yes
Page Range:pp. 471-496
Publisher:World Scientific
Keywords:Uncertainty in AI; explainable AI; non-parametric quantile estimation; order statistics; split conformal predictions
HGF - Research field:other
HGF - Program:other
HGF - Program Themes:other
DLR - Research area:no assignment
DLR - Program:no assignment
DLR - Research theme (Project):no assignment
Location: Bremerhaven
Institutes and Institutions:Institute for the Protection of Maritime Infrastructures > Maritime Security Technologies
Deposited By: Solano Carrillo, Edgardo
Deposited On:03 Aug 2023 15:08
Last Modified:19 Oct 2023 09:56

Repository Staff Only: item control page

Help & Contact
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.