Causal Discovery Under Local Privacy

Ruta Binkyte, Carlos Antonio Pinzón, Szilvia Lestyán, Kangsoo Jung, Héber Hwang Arcolezi, Catuscia Palamidessi
Proceedings of the Third Conference on Causal Learning and Reasoning, PMLR 236:325-383, 2024.

Abstract

Differential privacy is a widely adopted framework designed to safeguard the sensitive information of data providers within a data set. It is based on the application of controlled noise at the interface between the server that stores and processes the data, and the data consumers. Local differential privacy is a variant that allows data providers to apply the privatization mechanism themselves on their data individually. Therefore, it provides protection also in contexts in which the server, or even the data collector, cannot be trusted. The introduction of noise, however, inevitably affects the utility of the data, particularly by distorting the correlations between individual data components. This distortion can prove detrimental to tasks such as causal structure learning. In this paper, we consider various well-known locally differentially private mechanisms and compare the trade-off between the privacy they provide, and the accuracy of the causal structure produced by algorithms for causal learning when applied to data obfuscated by these mechanisms. Our analysis yields valuable insights for selecting appropriate local differentially private protocols for causal discovery tasks. We foresee that our findings will aid researchers and practitioners in conducting locally private causal discovery.

Cite this Paper

BibTeX
@InProceedings{pmlr-v236-binkyte24a, title = {Causal Discovery Under Local Privacy}, author = {Binkyte, Ruta and Pinz\'on, Carlos Antonio and Lesty\'an, Szilvia and Jung, Kangsoo and Arcolezi, H\'eber Hwang and Palamidessi, Catuscia}, booktitle = {Proceedings of the Third Conference on Causal Learning and Reasoning}, pages = {325--383}, year = {2024}, editor = {Locatello, Francesco and Didelez, Vanessa}, volume = {236}, series = {Proceedings of Machine Learning Research}, month = {01--03 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v236/binkyte24a/binkyte24a.pdf}, url = {https://proceedings.mlr.press/v236/binkyte24a.html}, abstract = {Differential privacy is a widely adopted framework designed to safeguard the sensitive information of data providers within a data set. It is based on the application of controlled noise at the interface between the server that stores and processes the data, and the data consumers. Local differential privacy is a variant that allows data providers to apply the privatization mechanism themselves on their data individually. Therefore, it provides protection also in contexts in which the server, or even the data collector, cannot be trusted. The introduction of noise, however, inevitably affects the utility of the data, particularly by distorting the correlations between individual data components. This distortion can prove detrimental to tasks such as causal structure learning. In this paper, we consider various well-known locally differentially private mechanisms and compare the trade-off between the privacy they provide, and the accuracy of the causal structure produced by algorithms for causal learning when applied to data obfuscated by these mechanisms. Our analysis yields valuable insights for selecting appropriate local differentially private protocols for causal discovery tasks. We foresee that our findings will aid researchers and practitioners in conducting locally private causal discovery.} }
Endnote
%0 Conference Paper %T Causal Discovery Under Local Privacy %A Ruta Binkyte %A Carlos Antonio Pinzón %A Szilvia Lestyán %A Kangsoo Jung %A Héber Hwang Arcolezi %A Catuscia Palamidessi %B Proceedings of the Third Conference on Causal Learning and Reasoning %C Proceedings of Machine Learning Research %D 2024 %E Francesco Locatello %E Vanessa Didelez %F pmlr-v236-binkyte24a %I PMLR %P 325--383 %U https://proceedings.mlr.press/v236/binkyte24a.html %V 236 %X Differential privacy is a widely adopted framework designed to safeguard the sensitive information of data providers within a data set. It is based on the application of controlled noise at the interface between the server that stores and processes the data, and the data consumers. Local differential privacy is a variant that allows data providers to apply the privatization mechanism themselves on their data individually. Therefore, it provides protection also in contexts in which the server, or even the data collector, cannot be trusted. The introduction of noise, however, inevitably affects the utility of the data, particularly by distorting the correlations between individual data components. This distortion can prove detrimental to tasks such as causal structure learning. In this paper, we consider various well-known locally differentially private mechanisms and compare the trade-off between the privacy they provide, and the accuracy of the causal structure produced by algorithms for causal learning when applied to data obfuscated by these mechanisms. Our analysis yields valuable insights for selecting appropriate local differentially private protocols for causal discovery tasks. We foresee that our findings will aid researchers and practitioners in conducting locally private causal discovery.
APA
Binkyte, R., Pinzón, C.A., Lestyán, S., Jung, K., Arcolezi, H.H. & Palamidessi, C.. (2024). Causal Discovery Under Local Privacy. Proceedings of the Third Conference on Causal Learning and Reasoning, in Proceedings of Machine Learning Research 236:325-383 Available from https://proceedings.mlr.press/v236/binkyte24a.html.

Related Material