SPPM: Sparse Privacy Preserving Mappings

Salman Salamatian, Nadia Fawaz Technicolor, Branislav Kveton Technicolor Labs, Nina Taft Technicolor
Proceedings of the 30th Conference on Uncertainty in Artificial Intelligence, PMLR R12:618-627, 2014.

Abstract

We study the problem of a user who has both public and private data, and wants to release the public data, e.g. to a recommendation service, yet simultaneously wants to protect his private data from being inferred via big data analytics. This problem has previously been formulated as a convex optimization problem with linear constraints where the objective is to minimize the mutual information between the private and released data. This attractive formulation faces a challenge in practice because when the underlying alphabet of the user profile is large, there are too many potential ways to distort the original profile. We address this fundamental scalability challenge. We propose to generate sparse privacy-preserving mappings by recasting the problem as a sequence of linear programs and solving each of these incrementally using an adaptation of Dantzig- Wolfe decomposition. We evaluate our approach on several datasets and demonstrate that nearly optimal privacy-preserving mappings can be learned quickly even at scale.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR12-salamatian14a, title = {{SPPM}: Sparse Privacy Preserving Mappings}, author = {Salamatian, Salman and Technicolor, Nadia Fawaz and Labs, Branislav Kveton Technicolor and Technicolor, Nina Taft}, booktitle = {Proceedings of the 30th Conference on Uncertainty in Artificial Intelligence}, pages = {618--627}, year = {2014}, editor = {Zhang, Nevin L. and Tian, Jin}, volume = {R12}, series = {Proceedings of Machine Learning Research}, month = {23--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r12/main/assets/salamatian14a/salamatian14a.pdf}, url = {https://proceedings.mlr.press/r12/salamatian14a.html}, abstract = {We study the problem of a user who has both public and private data, and wants to release the public data, e.g. to a recommendation service, yet simultaneously wants to protect his private data from being inferred via big data analytics. This problem has previously been formulated as a convex optimization problem with linear constraints where the objective is to minimize the mutual information between the private and released data. This attractive formulation faces a challenge in practice because when the underlying alphabet of the user profile is large, there are too many potential ways to distort the original profile. We address this fundamental scalability challenge. We propose to generate sparse privacy-preserving mappings by recasting the problem as a sequence of linear programs and solving each of these incrementally using an adaptation of Dantzig- Wolfe decomposition. We evaluate our approach on several datasets and demonstrate that nearly optimal privacy-preserving mappings can be learned quickly even at scale.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T SPPM: Sparse Privacy Preserving Mappings %A Salman Salamatian %A Nadia Fawaz Technicolor %A Branislav Kveton Technicolor Labs %A Nina Taft Technicolor %B Proceedings of the 30th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2014 %E Nevin L. Zhang %E Jin Tian %F pmlr-vR12-salamatian14a %I PMLR %P 618--627 %U https://proceedings.mlr.press/r12/salamatian14a.html %V R12 %X We study the problem of a user who has both public and private data, and wants to release the public data, e.g. to a recommendation service, yet simultaneously wants to protect his private data from being inferred via big data analytics. This problem has previously been formulated as a convex optimization problem with linear constraints where the objective is to minimize the mutual information between the private and released data. This attractive formulation faces a challenge in practice because when the underlying alphabet of the user profile is large, there are too many potential ways to distort the original profile. We address this fundamental scalability challenge. We propose to generate sparse privacy-preserving mappings by recasting the problem as a sequence of linear programs and solving each of these incrementally using an adaptation of Dantzig- Wolfe decomposition. We evaluate our approach on several datasets and demonstrate that nearly optimal privacy-preserving mappings can be learned quickly even at scale. %Z Reissued by PMLR on 04 October 2026.
APA
Salamatian, S., Technicolor, N.F., Labs, B.K.T. & Technicolor, N.T.. (2014). SPPM: Sparse Privacy Preserving Mappings. Proceedings of the 30th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R12:618-627 Available from https://proceedings.mlr.press/r12/salamatian14a.html. Reissued by PMLR on 04 October 2026.

Related Material