SETSUBUN: Revisiting Membership Inference Game for Evaluating Synthetic Data Generation

• Item type: Journal(1)
• 公開日: 2024-09-15
• タイトル: SETSUBUN: Revisiting Membership Inference Game for Evaluating Synthetic Data Generation
• タイトル / 言語: en
• タイトル: SETSUBUN: Revisiting Membership Inference Game for Evaluating Synthetic Data Generation
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: [特集:サプライチェーンを安全にするサイバーセキュリティ技術] synthetic data generation, membership inference, privacy protection, evaluation framework
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_6501
• 資源タイプ: journal article
• 著者所属: NTT Social Informatics Laboratories／Osaka University
• 著者所属: NTT Social Informatics Laboratories
• 著者所属: NTT Social Informatics Laboratories
• 著者所属: NTT Social Informatics Laboratories／Osaka University
• 著者所属: NTT Social Informatics Laboratories
• 著者所属: NTT Social Informatics Laboratories
• 著者所属: Osaka University
• 著者所属(英): en
• 著者所属(英): NTT Social Informatics Laboratories / Osaka University
• 著者所属(英): en
• 著者所属(英): NTT Social Informatics Laboratories
• 著者所属(英): en
• 著者所属(英): NTT Social Informatics Laboratories
• 著者所属(英): en
• 著者所属(英): NTT Social Informatics Laboratories / Osaka University
• 著者所属(英): en
• 著者所属(英): NTT Social Informatics Laboratories
• 著者所属(英): en
• 著者所属(英): NTT Social Informatics Laboratories
• 著者所属(英): en
• 著者所属(英): Osaka University
• 著者名: Takayuki, Miura; Masanobu, Kii; Toshiki, Shibahara; Kazuki, Iwahana; Tetsuya, Okuda; Atsunori, Ichikawa; Naoto, Yanai
• 著者名(英): Takayuki, Miura; Masanobu, Kii; Toshiki, Shibahara; Kazuki, Iwahana; Tetsuya, Okuda; Atsunori, Ichikawa; Naoto, Yanai
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: Synthetic data generation techniques are promising for anonymizing high-dimensional tabular datasets, and their privacy protection can be evaluated by membership inference attacks. However, the existing evaluation framework has limitations from two perspectives: (1) it cannot evaluate the worst-case because a target sample is chosen randomly; and (2) the decision criterion of an adversary's inference is black box since the adversary conducts membership inference by using machine learning models. In this paper, we propose a framework to overcome the above limitations in a simple and clear fashion. To cope with limitation (1), we introduce a statistical distance to choose a vulnerable target sample. To cope with limitation (2), we propose two interpretable inference methods. One is a method with typical statistics scores, and the other is a method with the Euclidean distance from the target sample. We conduct extensive experiments on two datasets and five synthesis algorithms to confirm the effectiveness of our framework. The experiments show that our framework enables us to evaluate privacy in synthetic data generation techniques more tightly.; ------------------------------; This is a preprint of an article intended for publication Journal of; Information Processing(JIP). This preprint should not be cited. This; article should be cited as: Journal of Information Processing Vol.32(2024) (online); DOI http://dx.doi.org/10.2197/ipsjjip.32.757; ------------------------------
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: Synthetic data generation techniques are promising for anonymizing high-dimensional tabular datasets, and their privacy protection can be evaluated by membership inference attacks. However, the existing evaluation framework has limitations from two perspectives: (1) it cannot evaluate the worst-case because a target sample is chosen randomly; and (2) the decision criterion of an adversary's inference is black box since the adversary conducts membership inference by using machine learning models. In this paper, we propose a framework to overcome the above limitations in a simple and clear fashion. To cope with limitation (1), we introduce a statistical distance to choose a vulnerable target sample. To cope with limitation (2), we propose two interpretable inference methods. One is a method with typical statistics scores, and the other is a method with the Euclidean distance from the target sample. We conduct extensive experiments on two datasets and five synthesis algorithms to confirm the effectiveness of our framework. The experiments show that our framework enables us to evaluate privacy in synthetic data generation techniques more tightly.; ------------------------------; This is a preprint of an article intended for publication Journal of; Information Processing(JIP). This preprint should not be cited. This; article should be cited as: Journal of Information Processing Vol.32(2024) (online); DOI http://dx.doi.org/10.2197/ipsjjip.32.757; ------------------------------
• 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: AN00116647
• 書誌情報: 情報処理学会論文誌; 巻 65, 号 9, 発行日 2024-09-15
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 1882-7764
• 公開者 / 言語: ja
• 公開者 / 出版者: 情報処理学会