PrIsing: Privacy-Preserving Peer Effect Estimation via Ising Model

Abhinav Chakraborty, Anirban Chatterjee, Abhinandan Dalal
Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, PMLR 238:2692-2700, 2024.

Abstract

The Ising model, originally developed as a spin-glass model for ferromagnetic elements, has gained popularity as a network-based model for capturing dependencies in agents’ outputs. Its increasing adoption in healthcare and the social sciences has raised privacy concerns regarding the confidentiality of agents’ responses. In this paper, we present a novel $(\varepsilon,\delta)$-differentially private algorithm specifically designed to protect the privacy of individual agents’ outcomes. Our algorithm allows for precise estimation of the natural parameter using a single network through an objective perturbation technique. Furthermore, we establish regret bounds for this algorithm and assess its performance on synthetic datasets and two real-world networks: one involving HIV status in a social network and the other concerning the political leaning of online blogs.

Cite this Paper


BibTeX
@InProceedings{pmlr-v238-chakraborty24a, title = {Pr{I}sing: Privacy-Preserving Peer Effect Estimation via {I}sing Model}, author = {Chakraborty, Abhinav and Chatterjee, Anirban and Dalal, Abhinandan}, booktitle = {Proceedings of The 27th International Conference on Artificial Intelligence and Statistics}, pages = {2692--2700}, year = {2024}, editor = {Dasgupta, Sanjoy and Mandt, Stephan and Li, Yingzhen}, volume = {238}, series = {Proceedings of Machine Learning Research}, month = {02--04 May}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v238/chakraborty24a/chakraborty24a.pdf}, url = {https://proceedings.mlr.press/v238/chakraborty24a.html}, abstract = {The Ising model, originally developed as a spin-glass model for ferromagnetic elements, has gained popularity as a network-based model for capturing dependencies in agents’ outputs. Its increasing adoption in healthcare and the social sciences has raised privacy concerns regarding the confidentiality of agents’ responses. In this paper, we present a novel $(\varepsilon,\delta)$-differentially private algorithm specifically designed to protect the privacy of individual agents’ outcomes. Our algorithm allows for precise estimation of the natural parameter using a single network through an objective perturbation technique. Furthermore, we establish regret bounds for this algorithm and assess its performance on synthetic datasets and two real-world networks: one involving HIV status in a social network and the other concerning the political leaning of online blogs.} }
Endnote
%0 Conference Paper %T PrIsing: Privacy-Preserving Peer Effect Estimation via Ising Model %A Abhinav Chakraborty %A Anirban Chatterjee %A Abhinandan Dalal %B Proceedings of The 27th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2024 %E Sanjoy Dasgupta %E Stephan Mandt %E Yingzhen Li %F pmlr-v238-chakraborty24a %I PMLR %P 2692--2700 %U https://proceedings.mlr.press/v238/chakraborty24a.html %V 238 %X The Ising model, originally developed as a spin-glass model for ferromagnetic elements, has gained popularity as a network-based model for capturing dependencies in agents’ outputs. Its increasing adoption in healthcare and the social sciences has raised privacy concerns regarding the confidentiality of agents’ responses. In this paper, we present a novel $(\varepsilon,\delta)$-differentially private algorithm specifically designed to protect the privacy of individual agents’ outcomes. Our algorithm allows for precise estimation of the natural parameter using a single network through an objective perturbation technique. Furthermore, we establish regret bounds for this algorithm and assess its performance on synthetic datasets and two real-world networks: one involving HIV status in a social network and the other concerning the political leaning of online blogs.
APA
Chakraborty, A., Chatterjee, A. & Dalal, A.. (2024). PrIsing: Privacy-Preserving Peer Effect Estimation via Ising Model. Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 238:2692-2700 Available from https://proceedings.mlr.press/v238/chakraborty24a.html.

Related Material