elismg

• Griesbauer, Elisabeth Maria; Czado, Claudia; Frigessi, Arnoldo & Haff, Ingrid Hobæk (2025). TVineSynth: A Truncated C-Vine Copula Generator of Synthetic Tabular Data to Balance Privacy and Utility. Proceedings of Machine Learning Research (PMLR). 258, s. 3511–3519. doi: https:/raw.githubusercontent.com/mlresearch/v258/main/assets/griesbauer25a/griesbauer25a.pdf. Fulltekst i vitenarkiv Vis sammendrag

  We propose TVineSynth, a vine copula based synthetic tabular data generator, which is designed to balance privacy and utility, using the vine tree structure and its truncation to do the trade-off. Contrary to synthetic data generators that achieve DP by globally adding noise, TVineSynth performs a controlled approximation of the estimated data generating distribution, so that it does not suffer from poor utility of the resulting synthetic data for downstream prediction tasks. TVineSynth introduces a targeted bias into the vine copula model that, combined with the specific tree structure of the vine, causes the model to zero out privacy-leaking dependencies while relying on those that are beneficial for utility. Privacy is here measured with membership (MIA) and attribute inference attacks (AIA). Further, we theoretically justify how the construction of TVineSynth ensures AIA privacy under a natural privacy measure for continuous sensitive attributes. When compared to competitor models, with and without DP, on simulated and on real-world data, TVineSynth achieves a superior privacy-utility balance.

Se alle arbeider i NVA

• Haff, Ingrid Hobæk; Griesbauer, Elisabeth Maria; Czado, Claudia & Frigessi, Arnoldo (2026). Synthetic data with vine copulas – balancing utility and privacy. Fulltekst i vitenarkiv
• Griesbauer, Elisabeth Maria; Czado, Claudia; Frigessi, Arnoldo & Haff, Ingrid Hobæk (2025). Synthetic data generation balancing privacy and utility, using vine copulas. Fulltekst i vitenarkiv Vis sammendrag

  The availability of high-quality data has led to tremendous advances in science, technology and society at large, when analysed by means of statistical and machine learning (ML) methods. However, real-world data in many cases cannot be made public to the research community due to privacy restrictions. This impairs progress especially in bio-medical research. Synthetic data can substitute the sensitive real data, as long as they do not disclose private aspects. This has proven to be successful in training downstream ML applications. We propose TVineSynth, a vine copula based synthetic tabular data generator. TVineSynth is designed to balance privacy and utility, using the vine tree structure and its truncation to do the trade-off. Contrary to synthetic data generators that achieve differential privacy (DP) by globally adding noise, TVineSynth performs a controlled approximation of the estimated data generating distribution. Because of this it does not suffer from poor utility of the resulting synthetic data for downstream prediction tasks. TVineSynth introduces a targeted bias into the vine copula model. Combined with the specific tree structure of the vine, this causes the model to zero out privacy-leaking dependencies while relying on those that are beneficial for utility. We theoretically justify how the construction of TVineSynth ensures privacy. When compared to competitor models, with and without DP, TVineSynth achieves a superior privacy-utility balance.

• Griesbauer, Elisabeth Maria; Czado, Claudia; Frigessi, Arnoldo & Haff, Ingrid Hobæk (2025). Poster: TVineSynth: A Truncated C-Vine Copula Generator of Synthetic Tabular Data to Balance Privacy and Utility. Fulltekst i vitenarkiv Vis sammendrag

  We propose TVineSynth, a vine copula based synthetic tabular data generator, which is designed to balance privacy and utility, using the vine tree structure and its truncation to do the trade-off. Contrary to synthetic data generators that achieve DP by globally adding noise, TVineSynth performs a controlled approximation of the estimated data generating distribution, so that it does not suffer from poor utility of the resulting synthetic data for downstream prediction tasks. TVineSynth introduces a targeted bias into the vine copula model that, combined with the specific tree structure of the vine, causes the model to zero out privacy-leaking dependencies while relying on those that are beneficial for utility. Privacy is here measured with membership (MIA) and attribute inference attacks (AIA). Further, we theoretically justify how the construction of TVineSynth ensures AIA privacy under a natural privacy measure for continuous sensitive attributes. When compared to competitor models, with and without DP, on simulated and on real-world data, TVineSynth achieves a superior privacy-utility balance.

• Griesbauer, Elisabeth Maria (2025). Beyond the real: Synthetic data explained. Fulltekst i vitenarkiv
• Griesbauer, Elisabeth Maria; Czado, Claudia; Frigessi, Arnoldo & Haff, Ingrid Hobæk (2025). Poster: TVineSynth: A Truncated C-Vine Copula Generator of Synthetic Tabular Data to Balance Privacy and Utility. Fulltekst i vitenarkiv Vis sammendrag

  We propose TVineSynth, a vine copula based synthetic tabular data generator, which is designed to balance privacy and utility, using the vine tree structure and its truncation to do the trade-off. Contrary to synthetic data generators that achieve DP by globally adding noise, TVineSynth performs a controlled approximation of the estimated data generating distribution, so that it does not suffer from poor utility of the resulting synthetic data for downstream prediction tasks. TVineSynth introduces a targeted bias into the vine copula model that, combined with the specific tree structure of the vine, causes the model to zero out privacy-leaking dependencies while relying on those that are beneficial for utility. Privacy is here measured with membership (MIA) and attribute inference attacks (AIA). Further, we theoretically justify how the construction of TVineSynth ensures AIA privacy under a natural privacy measure for continuous sensitive attributes. When compared to competitor models, with and without DP, on simulated and on real-world data, TVineSynth achieves a superior privacy-utility balance.

• Griesbauer, Elisabeth Maria (2025). Utover det virkelige: Rollen til syntetiske data i utviklingen av KI. Fulltekst i vitenarkiv Vis sammendrag

  Syntetiske data er kunstig genererte data laget ved hjelp av algoritmer, statistiske modeller og kunstig intelligens (KI), som Chat-GPT. Syntetiske data er utformet for å etterligne reelle data ved å gjengi deres statistiske egenskaper, men er ikke direkte kopier eller tilfeldig støy. Behovet for syntetiske data i KI-utvikling oppstår fra mangelen på tilgjengelige, rene, private og varierte reelle data av høy kvalitet, som er nødvendige for å trene KI-modeller. Syntetiske data kan utvide eksisterende datasett, sikre privacy, fjerne skjevheter og bidra til rettferdighet i KI-systemer. Imidlertid kan bruken av syntetiske data føre til "modellkollaps". Dette skjer når KI-modeller trenes mer på syntetiske data de selv genererer. Denne prosessen kan redusere modellens evne til å forstå og gjengi virkeligheten, ettersom de gradvis glemmer de opprinnelige reelle dataene. Derfor er det avgjørende at KI-modeller både trenes på reelle og syntetiske data for å sikre at læringen av KI-modeller opprettholdes over tid.

• Griesbauer, Elisabeth Maria; Czado, Claudia; Frigessi, Arnoldo & Haff, Ingrid Hobæk (2024). Synthetic data generation balancing privacy and utility, using vine copulas. doi: https:/nva.sikt.no/registration/0199812d9067-f1497889-e36d-4995-8686-7c68d1e99cdb/edit. Fulltekst i vitenarkiv Vis sammendrag

  The availability of diverse, high-quality data has led to tremendous advances in science, technology and society at large, when analysed by means of statistical and machine learning (ML) methods. However, real-world data, in many cases, cannot be made public to the research community due to privacy restrictions, obstructing progress, especially in bio- medical research. Synthetic data can substitute the sensitive real data, and as long as they do not disclose private aspects. This has proven to be successful in training downstream ML applications. We propose TVineSynth, a vine copula based synthetic tabular data generator, which is designed to balance privacy and utility, using the vine tree structure and its truncation to do the trade-off. Contrary to synthetic data generators that achieve differential privacy (DP) by globally adding noise, TVineSynth performs a controlled approximation of the estimated data generating distribution, so that it does not suffer from poor utility of the resulting synthetic data for downstream prediction tasks. TVineSynth introduces a targeted bias into the vine copula model that, combined with the specific tree structure of the vine, causes the model to zero out privacy-leaking dependencies while relying on those that are beneficial for utility. Privacy is here measured with membership (MIA) and attribute inference attacks (AIA). Further, we theoretically justify how the construction of TVineSynth ensures AIA privacy under a natural privacy measure for continuous sensitive attributes. When compa red to competitor models, with and without DP, on simulated and on real-world data, TVineSynth achieves a superior privacy-utility balance.

• Haff, Ingrid Hobæk; Griesbauer, Elisabeth Maria; Frigessi, Arnoldo & Czado, Claudia (2024). Synthetic data with vine copulas – balancing utility and privacy. Fulltekst i vitenarkiv
• Haff, Ingrid Hobæk; Griesbauer, Elisabeth Maria; Frigessi, Arnoldo & Czado, Claudia (2024). Synthetic data with vine copulas – balancing utility and privacy. Fulltekst i vitenarkiv
• Haff, Ingrid Hobæk; Griesbauer, Elisabeth Maria; Frigessi, Arnoldo & Czado, Claudia (2023). Synthetic data with vine-copulas - balancing utility and privacy. Fulltekst i vitenarkiv

Se alle arbeider i NVA