Dual Privacy Guarantees for Distributed Nash Equilibrium Seeking in Aggregative Games

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Meng, Qingtan, Ma, Qian
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910260600504320
author Meng, Qingtan
Ma, Qian
author_facet Meng, Qingtan
Ma, Qian
contents This paper investigates the privacy-preserving distributed Nash equilibrium seeking problem for aggregative games. A novel differential privacy mechanism is designed by incorporating stochastic event-triggering with stochastic quantization, which provides strong privacy protection by obfuscating the temporal patterns of information exchange among players and quantizing the transmitted information at triggering instants. Based on this mechanism, a differentially private distributed Nash equilibrium seeking algorithm with dual randomness is proposed. By embedding a decaying factor sequence into both the triggering condition and interaction terms among players, it is proved that the proposed algorithm can achieve rigorous $(0,δ)$-differential privacy at each iteration while maintaining provable convergence. Crucially, this privacy guarantee is sustained over infinite iterations for a sufficiently large quantization interval and a sufficiently small trigger threshold tuning coefficient. Moreover, the synergy between event-triggered communication and quantization significantly enhances communication efficiency. Simulation results verify the validity of the proposed approach.
format Preprint
id arxiv_https___arxiv_org_abs_2605_26931
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Dual Privacy Guarantees for Distributed Nash Equilibrium Seeking in Aggregative Games
Meng, Qingtan
Ma, Qian
Optimization and Control
This paper investigates the privacy-preserving distributed Nash equilibrium seeking problem for aggregative games. A novel differential privacy mechanism is designed by incorporating stochastic event-triggering with stochastic quantization, which provides strong privacy protection by obfuscating the temporal patterns of information exchange among players and quantizing the transmitted information at triggering instants. Based on this mechanism, a differentially private distributed Nash equilibrium seeking algorithm with dual randomness is proposed. By embedding a decaying factor sequence into both the triggering condition and interaction terms among players, it is proved that the proposed algorithm can achieve rigorous $(0,δ)$-differential privacy at each iteration while maintaining provable convergence. Crucially, this privacy guarantee is sustained over infinite iterations for a sufficiently large quantization interval and a sufficiently small trigger threshold tuning coefficient. Moreover, the synergy between event-triggered communication and quantization significantly enhances communication efficiency. Simulation results verify the validity of the proposed approach.
title Dual Privacy Guarantees for Distributed Nash Equilibrium Seeking in Aggregative Games
topic Optimization and Control
url https://arxiv.org/abs/2605.26931