Neural Criticality Metric for Object Detection Deep Neural Networks
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F22%3A43966110" target="_blank" >RIV/49777513:23520/22:43966110 - isvavai.cz</a>
Result on the web
<a href="https://link.springer.com/chapter/10.1007/978-3-031-14862-0_20" target="_blank" >https://link.springer.com/chapter/10.1007/978-3-031-14862-0_20</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1007/978-3-031-14862-0_20" target="_blank" >10.1007/978-3-031-14862-0_20</a>
Alternative languages
Result language
angličtina
Original language name
Neural Criticality Metric for Object Detection Deep Neural Networks
Original language description
The complexity of state-of-the-art Deep Neural Network (DNN) architectures exacerbates the search for safety relevant metrics and methods that could be used for functional safety assessments. In this article, we investigate Neurons' Criticality (the ability to affect the decision process) for several object detection DNN architectures. As a first step, we introduce the Neural Criticality metric for object detection DNNs and set a theoretical background. Subsequently, by conducting experiments, we verify that removing one neuron from the computational graph of a DNN can have a significant (positive, as well as negative) influence on the prediction's precision (object classification and localization). Finally, we build statistics for each neuron from pre-trained networks on the COCO object detection validation dataset and examine the network stability for the most critical neurons in order to prove our metric's validity.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
20205 - Automation and control systems
Result continuities
Project
<a href="/en/project/TN01000024" target="_blank" >TN01000024: National Competence Center - Cybernetics and Artificial Intelligence</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2022
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Computer Safety, Reliability and Security, SAFECOMP 2022 Workshops
ISBN
978-3-031-14861-3
ISSN
0302-9743
e-ISSN
1611-3349
Number of pages
13
Pages from-to
276-288
Publisher name
SPRINGER-VERLAG BERLIN
Place of publication
BERLIN
Event location
Munich, GERMANY
Event date
Sep 6, 2022
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
000866543800026