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

Using Predictive Uncertainty for Cleaning Noisy Annotations

Gütter, Jonas Aaron and Ulman, Hannah and Niebling, Julia (2022) Using Predictive Uncertainty for Cleaning Noisy Annotations. WAW Machine Learning 8, 2022-11-07 - 2022-11-09, Jena, Deutschland.

[img] PDF
513kB

Abstract

Predictive Uncertainty during model training can be used to assess wether a sample is correctly annotated or not. To see if this is also possible on remote sensing data, we applied the method on a building segmentation task.

Item URL in elib:https://elib.dlr.de/190240/
Document Type:Conference or Workshop Item (Poster)
Title:Using Predictive Uncertainty for Cleaning Noisy Annotations
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Gütter, Jonas AaronJonas.Guetter (at) dlr.deUNSPECIFIEDUNSPECIFIED
Ulman, Hannahhulman (at) princeton.eduUNSPECIFIEDUNSPECIFIED
Niebling, JuliaJulia.Niebling (at) dlr.deUNSPECIFIEDUNSPECIFIED
Date:November 2022
Refereed publication:No
Open Access:Yes
Gold Open Access:No
In SCOPUS:No
In ISI Web of Science:No
Status:Published
Keywords:Deep Learning, Remote Sensing, segmentation, label noise
Event Title:WAW Machine Learning 8
Event Location:Jena, Deutschland
Event Type:Workshop
Event Start Date:7 November 2022
Event End Date:9 November 2022
Organizer:DLR Institut für Datenwissenschaften Jena
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Space System Technology
DLR - Research area:Raumfahrt
DLR - Program:R SY - Space System Technology
DLR - Research theme (Project):R - Basic research in the field of machine learning
Location: Jena
Institutes and Institutions:Institute of Data Science
Institute of Data Science > Data Analysis and Intelligence
Deposited By: Gütter, Jonas Aaron
Deposited On:17 Nov 2022 15:28
Last Modified:15 Jan 2025 14:06

Repository Staff Only: item control page

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