Conference paper Open Access

# Explainable Deep Neural Networks for Multivariate Time Series Predictions

Assaf, Roy; Schumann, Anika

### DataCite XML Export

<?xml version='1.0' encoding='utf-8'?>
<identifier identifierType="URL">https://zenodo.org/record/3843273</identifier>
<creators>
<creator>
<creatorName>Assaf, Roy</creatorName>
<givenName>Roy</givenName>
<familyName>Assaf</familyName>
<affiliation>IBM Research Zurich</affiliation>
</creator>
<creator>
<creatorName>Schumann, Anika</creatorName>
<givenName>Anika</givenName>
<familyName>Schumann</familyName>
<affiliation>IBM Research Zurich</affiliation>
</creator>
</creators>
<titles>
<title>Explainable Deep Neural Networks for Multivariate Time Series Predictions</title>
</titles>
<publisher>Zenodo</publisher>
<publicationYear>2019</publicationYear>
<subjects>
<subject>AI: Machine Learning</subject>
<subject>AI: Knowledge Representation</subject>
<subject>Reasoning Applications: Energy</subject>
</subjects>
<dates>
<date dateType="Issued">2019-07-01</date>
</dates>
<resourceType resourceTypeGeneral="ConferencePaper"/>
<alternateIdentifiers>
<alternateIdentifier alternateIdentifierType="url">https://zenodo.org/record/3843273</alternateIdentifier>
</alternateIdentifiers>
<relatedIdentifiers>
<relatedIdentifier relatedIdentifierType="DOI" relationType="IsIdenticalTo">10.24963/ijcai.2019/932</relatedIdentifier>
</relatedIdentifiers>
<rightsList>
<rights rightsURI="info:eu-repo/semantics/openAccess">Open Access</rights>
</rightsList>
<descriptions>
<description descriptionType="Abstract">&lt;p&gt;We demonstrate that CNN deep neural networks can not only be used for making predictions based on multivariate time series data, but also for explaining these predictions. This is important for a number of applications where predictions are the basis for decisions and actions. Hence, confidence in the prediction result is crucial. We design a two stage convolutional neural network architecture which uses particular kernel sizes. This allows us to utilise gradient based techniques for generating saliency maps for both the time dimension and the features. These are then used for explaining which features during which time interval are responsible for a given prediction, as well as explaining during which time intervals was the joint contribution of all features most important for that prediction. We demonstrate our approach for predicting the average energy production of photovoltaic power plants and for explaining these predictions.&lt;/p&gt;</description>
</descriptions>
<fundingReferences>
<fundingReference>
<funderName>European Commission</funderName>
<funderIdentifier funderIdentifierType="Crossref Funder ID">10.13039/100010661</funderIdentifier>
<awardNumber awardURI="info:eu-repo/grantAgreement/EC/H2020/745625/">745625</awardNumber>
<awardTitle>Reliable OM decision tools and strategies for high LCoE reduction on Offshore wind</awardTitle>
</fundingReference>
</fundingReferences>
</resource>

143
105
views