<?xml version="1.0" encoding="utf-8"?>
<export-example>
  <doc>
    <id>789</id>
    <completedYear/>
    <publishedYear>2019</publishedYear>
    <thesisYearAccepted/>
    <language>deu</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>72</pageNumber>
    <edition/>
    <issue/>
    <volume/>
    <type>periodicalpart</type>
    <publisherName/>
    <publisherPlace>Regensburg</publisherPlace>
    <creatingCorporation>Ostbayerische Technische Hochschule Regensburg</creatingCorporation>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2019-06-30</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="deu">Forschung 2019</title>
    <abstract language="deu">Bericht mit Forschungsprojekten aus verschiedenen Bereichen der OTH Regensburg mit dem Schwerpunktthema "Künstliche Intelligenz" und einem Gespräch zur "Medizin der Zukunft"</abstract>
    <subTitle language="deu">Thema: Künstliche Intelligenz</subTitle>
    <identifier type="isbn">978-3-9818209-7-3</identifier>
    <identifier type="urn">urn:nbn:de:bvb:898-opus4-7890</identifier>
    <identifier type="doi">10.35096/othr/pub-789</identifier>
    <licence>Keine Lizenz - Es gilt das deutsche Urheberrecht: § 53 UrhG</licence>
    <author>Marie-Luise Appelhans</author>
    <author>Matthias Kampmann</author>
    <author>Jürgen Mottok</author>
    <author>Michael Riederer</author>
    <author>Klaus Nagl</author>
    <author>Oliver Steffens</author>
    <author>Jan Dünnweber</author>
    <author>Markus Wildgruber</author>
    <author>Julius Roth</author>
    <author>Timo Stadler</author>
    <author>Christoph Palm</author>
    <author>Martin Georg Weiß</author>
    <author>Sandra Rochholz</author>
    <author>Rudolf Bierl</author>
    <author>Andreas Gschossmann</author>
    <author>Sonja Haug</author>
    <author>Simon Schmidbauer</author>
    <author>Anna Koch</author>
    <author>Markus Westner</author>
    <author>Benedikt von Bary</author>
    <author>Andreas Ellermeier</author>
    <author>Daniel Vögele</author>
    <author>Frederik Maiwald</author>
    <author>Stefan Hierl</author>
    <author>Matthias Schlamp</author>
    <author>Ingo Ehrlich</author>
    <author>Marco Siegl</author>
    <author>Sven Hüntelmann</author>
    <author>Matthias Wildfeuer</author>
    <author>Oliver Brückl</author>
    <author>Michael Sterner</author>
    <author>Andreas Hofrichter</author>
    <author>Fabian Eckert</author>
    <author>Franz Bauer</author>
    <author>Belal Dawoud</author>
    <author>Hans-Peter Rabl</author>
    <author>Bernd Gamisch</author>
    <author>Ottfried Schmidt</author>
    <author>Michael Heberl</author>
    <author>Martin Thema</author>
    <author>Ulrike Mayer</author>
    <author>Johannes Eller</author>
    <author>Thomas Sippenauer</author>
    <author>Christian Adelt</author>
    <author>Matthias Haslbeck</author>
    <author>Bettina Vogl</author>
    <author>Wolfgang Mauerer</author>
    <author>Ralf Ramsauer</author>
    <author>Daniel Lohmann</author>
    <author>Irmengard Sax</author>
    <author>Thomas Gabor</author>
    <author>Sebastian Feld</author>
    <author>Claudia Linnhoff-Popien</author>
    <author>Robert Damian Ławrowski</author>
    <author>Christoph Langer</author>
    <author>Rupert Schreiner</author>
    <author>Josef Sellmair</author>
    <series>
      <title>Forschungsberichte der OTH Regensburg</title>
      <number>2019</number>
    </series>
    <subject>
      <language>deu</language>
      <type>swd</type>
      <value>Forschung</value>
    </subject>
    <subject>
      <language>deu</language>
      <type>swd</type>
      <value>Forschungsbericht</value>
    </subject>
    <subject>
      <language>deu</language>
      <type>swd</type>
      <value>Künstliche Intelligenz</value>
    </subject>
    <collection role="ddc" number="0">Informatik, Informationswissenschaft, allgemeine Werke</collection>
    <collection role="institutes" number="HL">Hochschulleitung/Hochschulverwaltung</collection>
    <collection role="othpublikationsherkunft" number="">Von der OTH Regensburg herausgegeben</collection>
    <collection role="institutes" number="IAFW">Zentrum für Forschung und Transfer (ZFT ab 2024; vorher: IAFW)</collection>
    <collection role="persons" number="sternerfenes">Sterner, Michael (Prof. Dr.) - FENES / Forschungsgruppe Energiespeicher</collection>
    <collection role="persons" number="bruecklfenes">Brückl, Oliver (Prof. Dr.) - FENES / Forschungsgruppe Energienetze</collection>
    <thesisPublisher>Ostbayerische Technische Hochschule Regensburg</thesisPublisher>
    <file>https://opus4.kobv.de/opus4-oth-regensburg/files/789/Forschungsbericht_2019.pdf</file>
  </doc>
  <doc>
    <id>858</id>
    <completedYear/>
    <publishedYear>2019</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>153</pageFirst>
    <pageLast>157</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <type>conferenceobject</type>
    <publisherName>VDE-Verlag</publisherName>
    <publisherPlace>Berlin</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2019-09-16</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">A Measure of Confidence of Artificial Neural Network Classifiers</title>
    <abstract language="eng">Confidence in results of an Artificial Neural Networks (ANNs) is increased by preferring to reject data, that is not trustful, instead of risking a misclassification. For this purpose a model is proposed that is able to recognize data, which differs significantly from the training data, during inference. The proposed model observes all activations of the hidden layers, as well as input and output layers of an ANN in a grey-box view. To make ANNs more robust in safety critical applications, this model can be used to reject flawed data, that is suspected to decrease the accuracy of the model. If this information is logged during inference, it can be used to improve the model, by training it specifically with the missing information. An experiment on the MNIST dataset is conducted and its results are discussed.</abstract>
    <parentTitle language="eng">ARCS Workshop 2019; 32nd International Conference on Architecture of Computing Systems,  20-21 May 2019, Copenhagen, Denmark</parentTitle>
    <identifier type="url">https://ieeexplore.ieee.org/document/8836211</identifier>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="BegutachtungStatus">peer-reviewed</enrichment>
    <author>Andreas Gschossmann</author>
    <author>Simon Jobst</author>
    <author>Jürgen Mottok</author>
    <author>Rudolf Bierl</author>
    <collection role="ddc" number="006">Spezielle Computerverfahren</collection>
    <collection role="institutes" number="FakANK">Fakultät Angewandte Natur- und Kulturwissenschaften</collection>
    <collection role="institutes" number="FakEI">Fakultät Elektro- und Informationstechnik</collection>
    <collection role="persons" number="bierlsappz">Bierl, Rudolf (Prof. Dr.) - Sensorik-ApplikationsZentrum</collection>
    <collection role="othforschungsschwerpunkt" number="16311">Digitalisierung</collection>
    <collection role="institutes" number="">Sensorik-Applikationszentrum (SappZ)</collection>
    <collection role="institutes" number="">Laboratory for Safe and Secure Systems (LAS3)</collection>
  </doc>
</export-example>
