Nothing but Regrets — Privacy-Preserving Federated Causal Discovery

Osman Mian, David Kaltenpoth, Michael Kamp, Jilles Vreeken
Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, PMLR 206:8263-8278, 2023.

Abstract

In critical applications, causal models are the prime choice for their trustworthiness and explainability. If data is inherently distributed and privacy-sensitive, federated learning allows for collaboratively training a joint model. Existing approaches for federated causal discovery share locally discovered causal model in every iteration, therewith not only revealing local structure but also leading to very high communication costs. Instead, we propose an approach for privacy-preserving federated causal discovery by distributed min-max regret optimization. We prove that max-regret is a consistent scoring criterion that can be used within the well-known Greedy Equivalence Search to discover causal networks in a federated setting and is provably privacy-preserving at the same time. Through extensive experiments, we show that our approach reliably discovers causal networks without ever looking at local data and beats the state of the art both in terms of the quality of discovered causal networks as well as communication efficiency.

Cite this Paper


BibTeX
@InProceedings{pmlr-v206-mian23a, title = {Nothing but Regrets — Privacy-Preserving Federated Causal Discovery}, author = {Mian, Osman and Kaltenpoth, David and Kamp, Michael and Vreeken, Jilles}, booktitle = {Proceedings of The 26th International Conference on Artificial Intelligence and Statistics}, pages = {8263--8278}, year = {2023}, editor = {Ruiz, Francisco and Dy, Jennifer and van de Meent, Jan-Willem}, volume = {206}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v206/mian23a/mian23a.pdf}, url = {https://proceedings.mlr.press/v206/mian23a.html}, abstract = {In critical applications, causal models are the prime choice for their trustworthiness and explainability. If data is inherently distributed and privacy-sensitive, federated learning allows for collaboratively training a joint model. Existing approaches for federated causal discovery share locally discovered causal model in every iteration, therewith not only revealing local structure but also leading to very high communication costs. Instead, we propose an approach for privacy-preserving federated causal discovery by distributed min-max regret optimization. We prove that max-regret is a consistent scoring criterion that can be used within the well-known Greedy Equivalence Search to discover causal networks in a federated setting and is provably privacy-preserving at the same time. Through extensive experiments, we show that our approach reliably discovers causal networks without ever looking at local data and beats the state of the art both in terms of the quality of discovered causal networks as well as communication efficiency.} }
Endnote
%0 Conference Paper %T Nothing but Regrets — Privacy-Preserving Federated Causal Discovery %A Osman Mian %A David Kaltenpoth %A Michael Kamp %A Jilles Vreeken %B Proceedings of The 26th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2023 %E Francisco Ruiz %E Jennifer Dy %E Jan-Willem van de Meent %F pmlr-v206-mian23a %I PMLR %P 8263--8278 %U https://proceedings.mlr.press/v206/mian23a.html %V 206 %X In critical applications, causal models are the prime choice for their trustworthiness and explainability. If data is inherently distributed and privacy-sensitive, federated learning allows for collaboratively training a joint model. Existing approaches for federated causal discovery share locally discovered causal model in every iteration, therewith not only revealing local structure but also leading to very high communication costs. Instead, we propose an approach for privacy-preserving federated causal discovery by distributed min-max regret optimization. We prove that max-regret is a consistent scoring criterion that can be used within the well-known Greedy Equivalence Search to discover causal networks in a federated setting and is provably privacy-preserving at the same time. Through extensive experiments, we show that our approach reliably discovers causal networks without ever looking at local data and beats the state of the art both in terms of the quality of discovered causal networks as well as communication efficiency.
APA
Mian, O., Kaltenpoth, D., Kamp, M. & Vreeken, J.. (2023). Nothing but Regrets — Privacy-Preserving Federated Causal Discovery. Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 206:8263-8278 Available from https://proceedings.mlr.press/v206/mian23a.html.

Related Material