JST CREST研究課題(自動運転セキュリティ)

Featured

JST CREST研究領域「基礎理論とシステム基盤技術の融合によるSociety 5.0のための基盤ソフトウェアの創出」に、研究課題「AI駆動型サイバーフィジカルシステムのセキュリティ評価・対策基盤」が採択されました。

自動運転システムを構成するセンサ、カメラ、AI、制御機構を対象とした敵対的入力の脅威を明らかにし、全体として攻撃に対する耐性を高めるための技術を開発し、有効性を評価します。本研究課題は主たる共同研究者として、東工大の佐久間淳先生(機械学習)、電気通信大学の菅原健先生(物理セキュリティ)、澤田賢治先生(制御システム)、慶應義塾大学の吉岡健太郎先生(センサ、システム回路)、UCIの佐藤貴海さん(自動運転セキュリティ)という、それぞれ異なるバックグランドを持つ大変強力なメンバにご参画頂きました。自動運転システムを構成する個別の要素技術に加え、それらを統合したEnd-to-Endのシステム全体を対象とし、理論、応用、実装を組み合わせた幅広いアプローチで研究を進めていきます。本研究課題で得られた知見や各種の artifact (データ、ソースコードなど)は、オープンソースの形式で公開していく予定です。

プロジェクトウェブサイト: https://crest.seclab.jp/

Received the JIP specially selected paper certificate!

We are pleased to announce that our paper submitted to JIP has received the JIP Specially Selected Paper certificate. This award is presented to authors whose work ranks in the top 10 percent of outstanding papers in each JIP issue. An earlier version of this paper was presented at NDSS 2023.

Kazuki Nomoto, Takuya Watanabe, Eitaro Shioji, Mitsuaki Akiyama, Tatsuya Mori, “Understanding the Inconsistencies in the Permissions Mechanism of Web Browsers,” Journal of Information Processing, 2023, Volume 31, Pages 620-642, September 2023 [PDF]

A new interesting paper published!

Our paper submitted to the IPSJ JIP has been published and is now available online. This study investigates the actual instances of fraudulent transfers in online banking services provided by an existing Japanese bank. Furthermore, we propose and evaluate detection methods for such activities. We believe that this research, targeting real-world services, is of significant value. Congratulations, Asomura-san!

Ichiro Asomura, Ryo Iijima, Tatsuya Mori, “Automating the Detection of Fraudulent Activities in Online Banking Service,” Journal of Information Processing, Volume 31, Pages 643-653, September 2023 [PDF]

A paper got accepted!

Our paper submitted to the Workshop on Privacy in the Electronic Society (WPES 2023) has been accepted for publication! In this study of 7,134 privacy policies, we identified significant differences between parent and subsidiary companies. We observed that as organizational complexity increased, policy similarity decreased. Congratulations, Mori-san!

Keika Mori, Yuta Takata, Daiki Ito, Masaki Kamizono, and Tatsuya Mori, “Impact Analysis of Organizational Structure of Group Companies on Privacy Policies,” Proceedings of he Workshop on Privacy in the Electronic Society (WPES 2023), November 2023

A paper got accepted!

Our paper submitted to Journal of Information Processing has been accepted for publication! In this study, we employed bidirectional encoder representations from transformers to assess 1,298 privacy and 139 security policies of Japanese companies. Results indicated over 90% covered user rights and security well, but lacked in data retention details. Despite recent law changes, many policies remain insufficiently updated. Congratulations, Mori-san!

Keika Mori, Tatsuya Nagai, Yuta Takata, Masaki Kamizono, and Tatsuya Mori, “Analysis of Privacy Compliance by Classifying Policies Before and After the Japanese Law Revision,” Journal of Information Processing, vol. xx, no. xx., pp. xxxx-xxxxx (in press)