Algebraic aggregation of random forests

Simple item page
dc.contributor.authorGossen, Frederik
dc.contributor.authorSteffen, Bernhard
dc.date.accessioned2022-03-10T15:02:50Z
dc.date.available2022-03-10T15:02:50Z
dc.date.issued2021-09-29
dc.description.abstractRandom Forests are one of the most popular classifiers in machine learning. The larger they are, the more precise the outcome of their predictions. However, this comes at a cost: it is increasingly difficult to understand why a Random Forest made a specific choice, and its running time for classification grows linearly with the size (number of trees). In this paper, we propose a method to aggregate large Random Forests into a single, semantically equivalent decision diagram which has the following two effects: (1) minimal, sufficient explanations for Random Forest-based classifications can be obtained by means of a simple three step reduction, and (2) the running time is radically improved. In fact, our experiments on various popular datasets show speed-ups of several orders of magnitude, while, at the same time, also significantly reducing the size of the required data structure.en
dc.identifier.urihttp://hdl.handle.net/2003/40778
dc.identifier.urihttp://dx.doi.org/10.17877/DE290R-22635
dc.language.isoende
dc.relation.ispartofseriesInternational journal on software tools for technology transfer;Vol. 23. 2021, Art. Nr. 635
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subjectRandom foresten
dc.subjectAlgebraic decision diagramen
dc.subjectAggregationen
dc.subjectExplainabilityen
dc.subjectInterpretabilityen
dc.subjectRunning time optimisationen
dc.subjectMemory optimisationen
dc.subject.ddc004
dc.subject.rswkEntscheidungsgraphde
dc.subject.rswkAggregationde
dc.subject.rswkLaufzeitde
dc.subject.rswkErklärungde
dc.subject.rswkKlassifikator <Informatik>de
dc.subject.rswkSpeicher <Informatik>de
dc.subject.rswkOptimierungde
dc.titleAlgebraic aggregation of random forestsen
dc.title.alternativetowards explainability and rapid evaluationen
dc.typeTextde
dc.type.publicationtypearticlede
dcterms.accessRightsopen access
eldorado.secondarypublicationtruede
eldorado.secondarypublication.primarycitationInternational journal on software tools for technology transfer. Vol. 23. 2021, Art. Nr. 635en
eldorado.secondarypublication.primaryidentifierhttps://doi.org/10.1007/s10009-021-00635-xde

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Gossen-Steffen2021_Article_AlgebraicAggregationOfRandomFo.pdf
Size:
1.49 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
4.85 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

LS 14 Software Engineering