A Preliminary Study on Pre-training for Improved Explainability in End-to-End Driving Models

• Item type: SIG Technical Reports(1)
• 公開日: 2024-01-18
• タイトル: A Preliminary Study on Pre-training for Improved Explainability in End-to-End Driving Models
• タイトル / 言語: en
• タイトル: A Preliminary Study on Pre-training for Improved Explainability in End-to-End Driving Models
• 言語: eng
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: Graduate School of Informatics, Nagoya University
• 著者所属: Graduate School of Informatics, Nagoya University
• 著者所属: Graduate School of Informatics, Nagoya University
• 著者所属: Graduate School of Informatics, Nagoya University
• 著者所属(英): en
• 著者所属(英): Graduate School of Informatics, Nagoya University
• 著者所属(英): en
• 著者所属(英): Graduate School of Informatics, Nagoya University
• 著者所属(英): en
• 著者所属(英): Graduate School of Informatics, Nagoya University
• 著者所属(英): en
• 著者所属(英): Graduate School of Informatics, Nagoya University
• 著者名: Chenkai, Zhang; Daisuke, Deguchi; Jialei, Chen; Hiroshi, Murase
• 著者名(英): Chenkai, Zhang; Daisuke, Deguchi; Jialei, Chen; Hiroshi, Murase
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: In the realm of autonomous driving, end-to-end models (E2EDMs) have gained prominence due to their high predictive accuracy. Such accuracy is attributed to the utilization of a backbone pre-trained on large datasets, subsequently fine-tuned on autonomous driving datasets. However, the inherent “black box” nature of these E2EDMs poses significant challenges in terms of explainability. Current methodologies predominantly focus on generating visual explanations for the E2EDMs’ decision-making process. Numerous approaches aim to enhance the explainability of these E2EDMs by fine-tuning with complicated architectures, supplemented by additional information, e.g., object position to develop more explainable E2EDMs. In this study, we diverge from the conventional approaches where significant effort is placed during the fine-tuning phase of E2EDMs. Our method focuses on training backbones before the fine-tuning phase. This preemptive strategy enables us to fine-tune more explainable E2EDMs without the need for additional information or complex training techniques.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: In the realm of autonomous driving, end-to-end models (E2EDMs) have gained prominence due to their high predictive accuracy. Such accuracy is attributed to the utilization of a backbone pre-trained on large datasets, subsequently fine-tuned on autonomous driving datasets. However, the inherent “black box” nature of these E2EDMs poses significant challenges in terms of explainability. Current methodologies predominantly focus on generating visual explanations for the E2EDMs’ decision-making process. Numerous approaches aim to enhance the explainability of these E2EDMs by fine-tuning with complicated architectures, supplemented by additional information, e.g., object position to develop more explainable E2EDMs. In this study, we diverge from the conventional approaches where significant effort is placed during the fine-tuning phase of E2EDMs. Our method focuses on training backbones before the fine-tuning phase. This preemptive strategy enables us to fine-tune more explainable E2EDMs without the need for additional information or complex training techniques.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AA11131797
• 書誌情報: 研究報告コンピュータビジョンとイメージメディア（CVIM）; 巻 2024-CVIM-236, 号 40, p. 1-4, 発行日 2024-01-18
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-8701
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会