We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: You Can't See the Forest for Its Trees: Assessing Deep Neural Network Testing via NeuraL Coverage

Abstract: This paper summarizes eight design requirements for DNN testing criteria, taking into account distribution properties and practical concerns. We then propose a new criterion, NLC, that satisfies all of these design requirements. NLC treats a single DNN layer as the basic computational unit (rather than a single neuron) and captures four critical features of neuron output distributions. Thus, NLC is denoted as NeuraL Coverage, which more accurately describes how neural networks comprehend inputs via approximated distributions rather than neurons. We demonstrate that NLC is significantly correlated with the diversity of a test suite across a number of tasks (classification and generation) and data formats (image and text). Its capacity to discover DNN prediction errors is promising. Test input mutation guided by NLC result in a greater quality and diversity of exposed erroneous behaviors.
Subjects: Machine Learning (cs.LG); Software Engineering (cs.SE)
Cite as: arXiv:2112.01955 [cs.LG]
  (or arXiv:2112.01955v1 [cs.LG] for this version)

Submission history

From: Yuanyuan Yuan [view email]
[v1] Fri, 3 Dec 2021 15:01:52 GMT (2725kb,D)

Link back to: arXiv, form interface, contact.