TableGAN-MCA: Evaluating Membership Collisions of GAN-Synthesized Tabular Data Releasing

Simple item page
dc.contributor.authorHu, A.
dc.contributor.authorXie, R.
dc.contributor.authorLu, Z.
dc.contributor.authorHu, A.
dc.contributor.authorXue, M.
dc.contributor.conferenceACM SIGSAC Conference on Computer and Communications Security (15 Nov 2021 - 19 Nov 2021 : Virtual Online (Republic of Korea))
dc.date.issued2021
dc.description.abstractGenerative Adversarial Networks (GAN)-synthesized table publishing lets people privately learn insights without access to the private table. However, existing studies on Membership Inference (MI) Attacks show promising results on disclosing membership of training datasets of GAN-synthesized tables. Different from those works focusing on discovering membership of a given data point, in this paper, we propose a novel Membership Collision Attack against GANs (TableGAN-MCA), which allows an adversary given only synthetic entries randomly sampled from a black-box generator to recover partial GAN training data. Namely, a GAN-synthesized table immune to state-of-the-art MI attacks is vulnerable to the TableGAN-MCA. The success of TableGAN-MCA is boosted by an observation that GAN-synthesized tables potentially collide with the training data of the generator. Our experimental evaluations on TableGAN-MCA have five main findings. First, TableGAN-MCA has a satisfying training data recovery rate on three commonly used real-world datasets against four generative models. Second, factors, including the size of GAN training data, GAN training epochs and the number of synthetic samples available to the adversary, are positively correlated to the success of TableGAN-MCA. Third, highly frequent data points have high risks of being recovered by TableGAN-MCA. Fourth, some unique data are exposed to unexpected high recovery risks in TableGAN-MCA, which may attribute to GAN’s generalization. Fifth, as expected, differential privacy, without the consideration of the correlations between features, does not show commendable mitigation effect against the TableGAN-MCA. Finally, we propose two mitigation methods and show promising privacy and utility trade-offs when protecting against TableGAN-MCA.
dc.description.statementofresponsibilityAoting Hu, Renjie Xie, Zhigang Lu, Aiqun Hu, Minhui Xue
dc.identifier.citationProceedings of the ACM Conference on Computer and Communications Security, 2021, pp.2096-2112
dc.identifier.doi10.1145/3460120.3485251
dc.identifier.isbn9781450384544
dc.identifier.issn1543-7221
dc.identifier.orcidXue, M. [0000-0001-5411-5039] [0000-0002-9172-4252]
dc.identifier.urihttps://hdl.handle.net/2440/135640
dc.language.isoen
dc.publisherAssociation for Computing Machinery (ACM)
dc.publisher.placeNew York, NY, United States
dc.relation.granthttp://purl.org/au-research/grants/arc/DP210102670
dc.rights© 2021 Association for Computing Machinery.
dc.source.urihttps://dl.acm.org/doi/proceedings/10.1145/3460120
dc.subjectSecurity and privacy; Computing methodologies; Machine learning
dc.titleTableGAN-MCA: Evaluating Membership Collisions of GAN-Synthesized Tabular Data Releasing
dc.typeConference paper
pubs.publication-statusPublished

Files

Collections

Computer Science publications