Antti Honkela

Shift-aware score transport for privacy auditing without retraining

Yaohong Yang, Gauri Pradhan, Qi Chen +2

Privacy auditing empirically assesses privacy leakage by using membership inference attacks to establish lower bounds on privacy loss. We consider an observational auditing setting…

Efficient Provably Private Classification with a Tabular Foundation Model

Talal Alrawajfeh, Cristiana Diaconu, Ossi Räisä +5

Tabular data underpin prediction and decision-making in medicine, finance, government and science, but often contain sensitive individual-level information, creating a need for acc…

On Choosing the $μ$ Parameter in Gaussian Differential Privacy

Bogdan Kulynych, Antti Honkela

Recent work argues for using Gaussian differential privacy (GDP) to report the privacy guarantees in privacy-preserving machine learning. We provide principled mappings from pure-D…

On Reliability of Membership Inference Vulnerability Evaluation

Joonas Jälkö, Gauri Pradhan, Ossi Räisä +1

Membership inference attacks (MIAs) are popular methods for empirically assessing the leakage of sensitive information in the training data through models or statistics learned fro…

Beyond Square Roots: A Memory-Efficient Explicit Factorization for Multi-Epoch Private Learning

Nikita P. Kalinin, Aki Rehn, Joel Daniel Andersson +2

Correlated-noise mechanisms are among the most promising approaches for improving the utility of differentially private model training, but rigorous guarantees require explicit, an…

Empirical Comparison of Membership Inference Attacks in Deep Transfer Learning

Yuxuan Bai, Gauri Pradhan, Marlon Tobaben +1

With the emergence of powerful large-scale foundation models, the training paradigm is increasingly shifting from from-scratch training to transfer learning. This enables high util…

On Optimal Hyperparameters for Differentially Private Deep Transfer Learning

Aki Rehn, Linzh Zhao, Mikko A. Heikkilä +1

Differentially private (DP) transfer learning, i.e., fine-tuning a pretrained model on private data, is the current state-of-the-art approach for training large models under privac…

An Interactive Framework for Finding the Optimal Trade-off in Differential Privacy

Yaohong Yang, Aki Rehn, Sammie Katt +2

Differential privacy (DP) is the standard for privacy-preserving analysis, and introduces a fundamental trade-off between privacy guarantees and model performance. Selecting the op…

Mitigating Disparate Impact of Differentially Private Learning through Bounded Adaptive Clipping

Linzh Zhao, Aki Rehn, Mikko A. Heikkilä +2

Differential privacy (DP) has become an essential framework for privacy-preserving machine learning. Existing DP learning methods, however, often have disparate impacts on model pr…

Gaussian DP for Reporting Differential Privacy Guarantees in Machine Learning

Juan Felipe Gomez, Bogdan Kulynych, Georgios Kaissis +4

Current practices for reporting differential privacy (DP) guarantees for machine learning (ML) algorithms such as DP-SGD provide an incomplete and potentially misleading picture. F…

Hyperparameters in Score-Based Membership Inference Attacks

Gauri Pradhan, Joonas Jälkö, Marlon Tobaben +1

Membership Inference Attacks (MIAs) have emerged as a valuable framework for evaluating privacy leakage by machine learning models. Score-based MIAs are distinguished, in particula…

Privacy Leakage via Output Label Space and Differentially Private Continual Learning

Marlon Tobaben, Talal Alrawajfeh, Marcus Klasson +3

Differential privacy (DP) is a formal privacy framework that enables training machine learning (ML) models while protecting individuals' data. As pointed out by prior work, ML mode…

NeurIPS 2023 Competition: Privacy Preserving Federated Learning Document VQA

Marlon Tobaben, Mohamed Ali Souibgui, Rubèn Tito +24

The Privacy Preserving Federated Learning Document VQA (PFL-DocVQA) competition challenged the community to develop provably private and communication-efficient solutions in a fede…

Efficient and Scalable Implementation of Differentially Private Deep Learning without Shortcuts

Sebastian Rodriguez Beltran, Marlon Tobaben, Joonas Jälkö +2

Differentially private stochastic gradient descent (DP-SGD) is the standard algorithm for training machine learning models under differential privacy (DP). The most common DP-SGD p…