Matches in SemOpenAlex for { <https://semopenalex.org/work/W2904298867> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W2904298867 abstract "Deep learning recently becomes popular because it brings significant improvements on a wide variety of classification and recognition tasks. However, with the population and increasing usage of deep learning based models, not many people take into account the potential security risks which are likely to cause accidents in them. This paper mainly studies on the potential safety hazards in the obstacle recognition and processing system (ORPS) of the self-driving cars, which is constructed by deep learning architecture. We perform an attack that embeds a backdoor in the Mask R-CNN in ORPS by poisoning the dataset. The experiment result shows that it is possible to embed a backdoor in ORPS. We can see that the backdoored network can accurately recognize and trigger the backdoors in the poisoned dataset, which obviously change the size of bounding box and corresponding mask of those poisoned instances. But on the other hand, embedding a backdoor in the deep learning based model will only slightly affect the accuracy of detecting objects without backdoor triggers, which is imperceptible for users. Furthermore, in order to study the working mode of the backdoor and the possibility of detecting the backdoor in the network, we visualize the weights matrices in the backdoored network and try to modify them, but the results show that the existence of the backdoor in network is very cryptic, so it is difficult for users to detect and filter it. Eventually, we hope that our simple work can arouse people’s attention to the self-driving technology and even other deep learning based models." @default.
- W2904298867 created "2018-12-22" @default.
- W2904298867 creator A5028769645 @default.
- W2904298867 creator A5036385088 @default.
- W2904298867 creator A5040240016 @default.
- W2904298867 date "2018-01-01" @default.
- W2904298867 modified "2023-10-06" @default.
- W2904298867 title "Generating Misleading Labels in Machine Learning Models" @default.
- W2904298867 cites W1849277567 @default.
- W2904298867 cites W1903029394 @default.
- W2904298867 cites W1977177161 @default.
- W2904298867 cites W1978736542 @default.
- W2904298867 cites W2031489346 @default.
- W2904298867 cites W2100495367 @default.
- W2904298867 cites W2102605133 @default.
- W2904298867 cites W2108598243 @default.
- W2904298867 cites W2112796928 @default.
- W2904298867 cites W2147800946 @default.
- W2904298867 cites W2165698076 @default.
- W2904298867 cites W2179352600 @default.
- W2904298867 cites W2194775991 @default.
- W2904298867 cites W2340897893 @default.
- W2904298867 cites W2474389331 @default.
- W2904298867 cites W2543927648 @default.
- W2904298867 cites W2603766943 @default.
- W2904298867 cites W2753783305 @default.
- W2904298867 cites W2963150697 @default.
- W2904298867 doi "https://doi.org/10.1007/978-3-030-05054-2_12" @default.
- W2904298867 hasPublicationYear "2018" @default.
- W2904298867 type Work @default.
- W2904298867 sameAs 2904298867 @default.
- W2904298867 citedByCount "0" @default.
- W2904298867 crossrefType "book-chapter" @default.
- W2904298867 hasAuthorship W2904298867A5028769645 @default.
- W2904298867 hasAuthorship W2904298867A5036385088 @default.
- W2904298867 hasAuthorship W2904298867A5040240016 @default.
- W2904298867 hasConcept C108583219 @default.
- W2904298867 hasConcept C119857082 @default.
- W2904298867 hasConcept C154945302 @default.
- W2904298867 hasConcept C2781045450 @default.
- W2904298867 hasConcept C38652104 @default.
- W2904298867 hasConcept C41008148 @default.
- W2904298867 hasConcept C63584917 @default.
- W2904298867 hasConceptScore W2904298867C108583219 @default.
- W2904298867 hasConceptScore W2904298867C119857082 @default.
- W2904298867 hasConceptScore W2904298867C154945302 @default.
- W2904298867 hasConceptScore W2904298867C2781045450 @default.
- W2904298867 hasConceptScore W2904298867C38652104 @default.
- W2904298867 hasConceptScore W2904298867C41008148 @default.
- W2904298867 hasConceptScore W2904298867C63584917 @default.
- W2904298867 hasLocation W29042988671 @default.
- W2904298867 hasOpenAccess W2904298867 @default.
- W2904298867 hasPrimaryLocation W29042988671 @default.
- W2904298867 hasRelatedWork W2889233174 @default.
- W2904298867 hasRelatedWork W2900018096 @default.
- W2904298867 hasRelatedWork W2901221722 @default.
- W2904298867 hasRelatedWork W2969596189 @default.
- W2904298867 hasRelatedWork W2972120770 @default.
- W2904298867 hasRelatedWork W2985913519 @default.
- W2904298867 hasRelatedWork W2993846550 @default.
- W2904298867 hasRelatedWork W3009072493 @default.
- W2904298867 hasRelatedWork W3037928287 @default.
- W2904298867 hasRelatedWork W3041668106 @default.
- W2904298867 hasRelatedWork W3086120435 @default.
- W2904298867 hasRelatedWork W3106047871 @default.
- W2904298867 hasRelatedWork W3140988292 @default.
- W2904298867 hasRelatedWork W3167794394 @default.
- W2904298867 hasRelatedWork W3175215793 @default.
- W2904298867 hasRelatedWork W3182911748 @default.
- W2904298867 hasRelatedWork W3188624435 @default.
- W2904298867 hasRelatedWork W3199529221 @default.
- W2904298867 hasRelatedWork W3203050978 @default.
- W2904298867 hasRelatedWork W3205484173 @default.
- W2904298867 isParatext "false" @default.
- W2904298867 isRetracted "false" @default.
- W2904298867 magId "2904298867" @default.
- W2904298867 workType "book-chapter" @default.