Pith. sign in

REVIEW 3 cited by

Comprehensive Privacy Analysis of Deep Learning: Passive and Active White-box Inference Attacks against Centralized and Federated Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1812.00910 v2 pith:633F5YUT submitted 2018-12-03 stat.ML cs.CRcs.LG

Comprehensive Privacy Analysis of Deep Learning: Passive and Active White-box Inference Attacks against Centralized and Federated Learning

classification stat.ML cs.CRcs.LG
keywords inferenceattackslearningdeepmodelswhite-boxprivacytraining
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Deep neural networks are susceptible to various inference attacks as they remember information about their training data. We design white-box inference attacks to perform a comprehensive privacy analysis of deep learning models. We measure the privacy leakage through parameters of fully trained models as well as the parameter updates of models during training. We design inference algorithms for both centralized and federated learning, with respect to passive and active inference attackers, and assuming different adversary prior knowledge. We evaluate our novel white-box membership inference attacks against deep learning algorithms to trace their training data records. We show that a straightforward extension of the known black-box attacks to the white-box setting (through analyzing the outputs of activation functions) is ineffective. We therefore design new algorithms tailored to the white-box setting by exploiting the privacy vulnerabilities of the stochastic gradient descent algorithm, which is the algorithm used to train deep neural networks. We investigate the reasons why deep learning models may leak information about their training data. We then show that even well-generalized models are significantly susceptible to white-box membership inference attacks, by analyzing state-of-the-art pre-trained and publicly available models for the CIFAR dataset. We also show how adversarial participants, in the federated learning setting, can successfully run active membership inference attacks against other participants, even when the global model achieves high prediction accuracies.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Towards Characterizing and Limiting Information Exposure in DNN Layers

    cs.CR 2019-07 unverdicted novelty 6.0

    Framework quantifies per-layer sensitive information exposure in DNNs via generalization error and evaluates TEE-based protection for the most exposed layers against white-box membership inference.

  2. Code Division Modulation Layers Against Forgetting and Inference in Continual Gait Identification

    cs.MM 2026-07 conditional novelty 5.0

    A fixed random sign-pattern layer per task preserves continual-learning accuracy and blocks confidence-based membership inference for attackers who lack the code.

  3. Revisiting Privacy Preservation in Brain-Computer Interfaces: Conceptual Boundaries, Risk Pathways, and a Protection-Strength Grading Framework

    cs.AI 2026-05 unverdicted novelty 4.0

    The paper synthesizes BCI privacy risks and introduces a three-dimensional framework that grades existing protection methods into four strength levels while flagging mental privacy as an unresolved neuroethical issue.