Please use this identifier to cite or link to this item:
For citation please use:
Main Title: ML2Motif—Reliable extraction of discriminative sequence motifs from learning machines
Author(s): Vidovic, Marina Marie-Claire
Kloft, Marius
Müller, Klaus-Robert
Görnitz, Nico
Type: Article
Abstract: High prediction accuracies are not the only objective to consider when solving problems using machine learning. Instead, particular scientific applications require some explanation of the learned prediction function. For computational biology, positional oligomer importance matrices (POIMs) have been successfully applied to explain the decision of support vector machines (SVMs) using weighted-degree (WD) kernels. To extract relevant biological motifs from POIMs, the motifPOIM method has been devised and showed promising results on real-world data. Our contribution in this paper is twofold: as an extension to POIMs, we propose gPOIM, a general measure of feature importance for arbitrary learning machines and feature sets (including, but not limited to, SVMs and CNNs) and devise a sampling strategy for efficient computation. As a second contribution, we derive a convex formulation of motifPOIMs that leads to more reliable motif extraction from gPOIMs. Empirical evaluations confirm the usefulness of our approach on artificially generated data as well as on real-world datasets.
Subject(s): neural network
machine learning
computational biology
non-linear learning machines
Issue Date: 27-Mar-2017
Date Available: 26-Nov-2020
Language Code: en
DDC Class: 006 Spezielle Computerverfahren
Sponsor/Funder: BMBF, 01IB15001B, ALICE II - Autonomes Lernen in komplexen Umgebungen 2 (Autonomous Learning in Complex Environments 2)
BMBF, 031L0023A, PREDICT - Umfassende Datenintegration zur Verbesserung onkologischer Therapien - Teilprojekt A
BMBF, 031B0187B, Zuchtwert Mustererkennung in Hybridkulturarten (BreedPatH)-Teilprojekt B
BMBF, 01IS14013A, Verbundprojekt: BBDC - Berliner Kompetenzzentrum für Big Data
Journal Title: PLOS ONE
Publisher: Public Library of Science (PLOS)
Volume: 12
Issue: 3
Article Number: e0174392
Publisher DOI: 10.1371/journal.pone.0174392
EISSN: 1932-6203
TU Affiliation(s): Fak. 4 Elektrotechnik und Informatik » Inst. Softwaretechnik und Theoretische Informatik » FG Maschinelles Lernen
Appears in Collections:Technische Universität Berlin » Publications

Files in This Item:
Format: Adobe PDF | Size: 5.59 MB
DownloadShow Preview

Item Export Bar

This item is licensed under a Creative Commons License Creative Commons