Neural Criticality Metric for Object Detection Deep Neural Networks

Kód výsledku v 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>

Výsledek na webu

<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>

Jazyk výsledku

angličtina

Název v původním jazyce

Neural Criticality Metric for Object Detection Deep Neural Networks

Popis výsledku v původním jazyce

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&apos; 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&apos;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&apos;s validity.

Název v anglickém jazyce

Neural Criticality Metric for Object Detection Deep Neural Networks

Popis výsledku anglicky

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&apos; 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&apos;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&apos;s validity.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20205 - Automation and control systems

Projekt

<a href="/cs/project/TN01000024" target="_blank" >TN01000024: Národní centrum kompetence - Kybernetika a umělá inteligence</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2022

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Computer Safety, Reliability and Security, SAFECOMP 2022 Workshops

ISBN

978-3-031-14861-3

ISSN

0302-9743

e-ISSN

1611-3349

Počet stran výsledku

13

Strana od-do

276-288

Název nakladatele

SPRINGER-VERLAG BERLIN

Místo vydání

BERLIN

Místo konání akce

Munich, GERMANY

Datum konání akce

6. 9. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000866543800026