Cloud-Enabled Differentially Private Multiagent Optimization With Constraints

We present an optimization framework for solving multiagent convex programs subject to inequality constraints while keeping the agents' state trajectories private. Each agent has an objective function depending only upon its own state and the agents are collectively subject to global constraint...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on control of network systems Vol. 5; no. 4; pp. 1693 - 1706
Main Authors: Hale, Matthew T., Egerstedt, Magnus
Format: Journal Article
Language:English
Published: Piscataway IEEE 01.12.2018
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Algorithms
Cloud computing
Computer simulation
Convergence
Data privacy
Decentralized control
Eavesdropping
Linear programming
Multiagent systems
Networked control systems
Optimization
Privacy
Trajectories
Trajectory
ISSN:2325-5870, 2372-2533
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present an optimization framework for solving multiagent convex programs subject to inequality constraints while keeping the agents' state trajectories private. Each agent has an objective function depending only upon its own state and the agents are collectively subject to global constraints. The agents do not directly communicate with each other but instead route messages through a trusted cloud computer. The cloud adds noise to data being sent to the agents in accordance with the framework of differential privacy and, thus, keeps each agent's state trajectory private from all other agents and any eavesdroppers. This private problem can be viewed as a stochastic variational inequality, and it is solved using a projection-based method for solving variational inequalities that resemble a noisy primal-dual gradient algorithm. Convergence of the optimization algorithm in the presence of noise is proven, and a quantifiable tradeoff between privacy and convergence is extracted from this proof. Simulation results are provided that demonstrate numerical convergence for both <inline-formula><tex-math notation="LaTeX">\epsilon</tex-math></inline-formula>-differential privacy and <inline-formula><tex-math notation="LaTeX">(\epsilon, \delta)</tex-math></inline-formula>-differential privacy.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2325-5870
2372-2533
DOI:10.1109/TCNS.2017.2751458