國立虎尾科技大學 |

A Framework for Generating Dangerous Scenes : = Towards Explaining Realistic Driving Trajectories.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	A Framework for Generating Dangerous Scenes :/
Reminder of title:	Towards Explaining Realistic Driving Trajectories.
Author:	Xu, Shengjie.
Description:	1 online resource (84 pages)
Notes:	Source: Masters Abstracts International, Volume: 85-02.
Contained By:	Masters Abstracts International85-02.
Subject:	Electrical engineering. -
Online resource:	click for full text (PQDT)
ISBN:	9798380068499

A Framework for Generating Dangerous Scenes : = Towards Explaining Realistic Driving Trajectories.
Xu, Shengjie.

A Framework for Generating Dangerous Scenes :Towards Explaining Realistic Driving Trajectories. - 1 online resource (84 pages)

Source: Masters Abstracts International, Volume: 85-02.

Thesis (M.S.)--University of California, Santa Cruz, 2023.

Includes bibliographical references

Deep neural networks are black box models that are hard to interpret by humans. However, organizations developing AI models must ensure transparency and accountability by providing the public with a comprehensive understanding of model functionality. We suggest integrating explainability information as feedback during the development, verification, and testing of models. Our testing framework provides the following insight during the neural network training: Does the model equally effective for minor variations in the input data? In this thesis, we showed the explainability differences by comparing original and altered autonomous driving datasets for neural network training and explainability. We propose a framework for perturbing autonomous vehicle datasets, the DANGER framework, which generates edge-case images on top of current autonomous driving datasets. The inputs to DANGER are photorealistic datasets from real driving scenarios. We present the DANGER algorithm for vehicle position manipulation and the interface towards the renderer module and present five scenario-level dangerous primitives generation applied to the virtual KITTI, virtual KITTI 2, and Waymo datasets. We contribute two innovations in our study: (a) Our experiments prove that DANGER can be used as a framework for expanding the current datasets to cover generative while realistic and anomalous corner cases; (b) We tested the feasibility of providing interpretable information feedback in a generic deep neural network training by providing the Grad-CAM instability level.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798380068499Subjects--Topical Terms:

596380
Electrical engineering.
Subjects--Index Terms:

Autonomous vehicleIndex Terms--Genre/Form:

554714
Electronic books.

A Framework for Generating Dangerous Scenes : = Towards Explaining Realistic Driving Trajectories.
LDR:02960ntm a22003977 4500 001 1146228
005 20240812064346.5
006 m o d
007 cr bn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798380068499
035 $a (MiAaPQ)AAI29059974
035 $a AAI29059974
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Xu, Shengjie. $3 1471585
245 1 2 $a A Framework for Generating Dangerous Scenes : $b Towards Explaining Realistic Driving Trajectories.
264 0 $c 2023
300 $a 1 online resource (84 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 85-02.
500 $a Advisor: Teodorescu, Mircea.
502 $a Thesis (M.S.)--University of California, Santa Cruz, 2023.
504 $a Includes bibliographical references
520 $a Deep neural networks are black box models that are hard to interpret by humans. However, organizations developing AI models must ensure transparency and accountability by providing the public with a comprehensive understanding of model functionality. We suggest integrating explainability information as feedback during the development, verification, and testing of models. Our testing framework provides the following insight during the neural network training: Does the model equally effective for minor variations in the input data? In this thesis, we showed the explainability differences by comparing original and altered autonomous driving datasets for neural network training and explainability. We propose a framework for perturbing autonomous vehicle datasets, the DANGER framework, which generates edge-case images on top of current autonomous driving datasets. The inputs to DANGER are photorealistic datasets from real driving scenarios. We present the DANGER algorithm for vehicle position manipulation and the interface towards the renderer module and present five scenario-level dangerous primitives generation applied to the virtual KITTI, virtual KITTI 2, and Waymo datasets. We contribute two innovations in our study: (a) Our experiments prove that DANGER can be used as a framework for expanding the current datasets to cover generative while realistic and anomalous corner cases; (b) We tested the feasibility of providing interpretable information feedback in a generic deep neural network training by providing the Grad-CAM instability level.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Electrical engineering. $3 596380
650 4 $a Computer science. $3 573171
653 $a Autonomous vehicle
653 $a Generative AI
653 $a Neural network training
653 $a Black box models
653 $a Model functionality
655 7 $a Electronic books. $2 local $3 554714
690 $a 0544
690 $a 0984
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a University of California, Santa Cruz. $b Electrical Engineering. $3 1148555
773 0 $t Masters Abstracts International $g 85-02.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29059974 $z click for full text (PQDT)