Conference paper Open Access

INPUT COMPLEXITY AND OUT-OF-DISTRIBUTION DETECTION WITH LIKELIHOOD-BASED GENERATIVE MODELS

Serra, Joan; Álvarez, David; Gómez, Vicenç; Slizovskaia , Olga; F. Núñez, José; Luque, Jordi


Dublin Core Export

<?xml version='1.0' encoding='utf-8'?>
<oai_dc:dc xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
  <dc:creator>Serra, Joan</dc:creator>
  <dc:creator>Álvarez, David</dc:creator>
  <dc:creator>Gómez, Vicenç</dc:creator>
  <dc:creator>Slizovskaia , Olga</dc:creator>
  <dc:creator>F. Núñez, José</dc:creator>
  <dc:creator>Luque, Jordi</dc:creator>
  <dc:date>2020-04-30</dc:date>
  <dc:description>Likelihood-based generative models are a promising resource to detect out-of- distribution (OOD) inputs which could compromise the robustness or reliability of a machine learning system. However, likelihoods derived from such models have been shown to be problematic for detecting certain types of inputs that sig- nificantly differ from training data. In this paper, we pose that this problem is due to the excessive influence that input complexity has in generative models’ likelihoods. We report a set of experiments supporting this hypothesis, and use an estimate of input complexity to derive an efficient and parameter-free OOD score, which can be seen as a likelihood-ratio, akin to Bayesian model compari- son. We find such score to perform comparably to, or even better than, existing OOD detection approaches under a wide range of data sets, models, model sizes, and complexity estimates.</dc:description>
  <dc:description>Accepted for ICLR2020</dc:description>
  <dc:identifier>https://zenodo.org/record/5253740</dc:identifier>
  <dc:identifier>10.5281/zenodo.5253740</dc:identifier>
  <dc:identifier>oai:zenodo.org:5253740</dc:identifier>
  <dc:relation>info:eu-repo/grantAgreement/EC/Horizon 2020 Framework Programme - Research and Innovation action/871793/</dc:relation>
  <dc:relation>doi:10.5281/zenodo.5253739</dc:relation>
  <dc:relation>url:https://zenodo.org/communities/accordion</dc:relation>
  <dc:rights>info:eu-repo/semantics/openAccess</dc:rights>
  <dc:rights>https://creativecommons.org/licenses/by/4.0/legalcode</dc:rights>
  <dc:title>INPUT COMPLEXITY AND OUT-OF-DISTRIBUTION DETECTION WITH LIKELIHOOD-BASED GENERATIVE MODELS</dc:title>
  <dc:type>info:eu-repo/semantics/conferencePaper</dc:type>
  <dc:type>publication-conferencepaper</dc:type>
</oai_dc:dc>
17
12
views
downloads
All versions This version
Views 1717
Downloads 1212
Data volume 23.3 MB23.3 MB
Unique views 1414
Unique downloads 1212

Share

Cite as