6 papers
Accuracy-First Rényi Differential Privacy and Post-Processing Immunity
Ossi Räisä, Antti Koskela, Antti Honkela
The accuracy-first perspective of differential privacy addresses an important shortcoming by allowing a data analyst to adaptively adjust the quantitative privacy bound instead of…
On Reliability of Efficient 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…
-Differential Privacy Filters: Validity and Approximate Solutions
Long Tran, Antti Koskela, Ossi Räisä +1
Accounting for privacy loss under fully adaptive composition -- where mechanism choice and privacy parameters may depend on the history of prior outputs -- is a central challenge i…
Position: All Current Generative Fidelity and Diversity Metrics are Flawed
Ossi Räisä, Boris van Breugel, Mihaela van der Schaar
Any method's development and practical application is limited by our ability to measure its reliability. The popularity of generative modeling emphasizes the importance of good syn…
Noise-Aware Differentially Private Regression via Meta-Learning
Ossi Räisä, Stratis Markou, Matthew Ashman +4
Many high-stakes applications require machine learning models that protect user privacy and provide well-calibrated, accurate predictions. While Differential Privacy (DP) is the go…
A Bias-Variance Decomposition for Ensembles over Multiple Synthetic Datasets
Ossi Räisä, Antti Honkela
Recent studies have highlighted the benefits of generating multiple synthetic datasets for supervised learning, from increased accuracy to more effective model selection and uncert…