Dynamics based privacy preservation in decentralized optimization
From MaRDI portal
Abstract: With decentralized optimization having increased applications in various domains ranging from machine learning, control, sensor networks, to robotics, its privacy is also receiving increased attention. Existing privacy-preserving approaches for decentralized optimization achieve privacy preservation by patching decentralized optimization with information-technology privacy mechanisms such as differential privacy or homomorphic encryption, which either sacrifices optimization accuracy or incurs heavy computation/communication overhead. We propose an inherently privacy-preserving decentralized optimization algorithm by exploiting the robustness of decentralized optimization to uncertainties in optimization dynamics. More specifically, we present a general decentralized optimization framework, based on which we show that privacy can be enabled in decentralized optimization by adding randomness in optimization parameters. We further show that the added randomness has no influence on the accuracy of optimization, and prove that our inherently privacy-preserving algorithm has -linear convergence when the global objective function is smooth and strongly convex. We also rigorously prove that the proposed algorithm can avoid the gradient of a node from being inferable by other nodes. Numerical simulation results confirm the theoretical predictions.
Recommendations
- Enabling Privacy-Preservation in Decentralized Optimization
- Decentralized Stochastic Optimization With Inherent Privacy Protection
- Quantization Enabled Privacy Protection in Decentralized Stochastic Optimization
- Privacy preserving distributed optimization using homomorphic encryption
- Privacy-Preserving Dual Averaging With Arbitrary Initial Conditions for Distributed Optimization
- Distributed Computing – IWDC 2005
- Decentralized nonconvex optimization with guaranteed privacy and accuracy
- Privacy-Preserving Distributed Optimization via Subspace Perturbation: A General Framework
- Differentially Private Distributed Constrained Optimization
- A privacy-preserving method to optimize distributed resource allocation
Cites work
- Achieving Geometric Convergence for Distributed Optimization Over Time-Varying Graphs
- ADD-OPT: Accelerated Distributed Directed Optimization
- Analysis of optimization algorithms via integral quadratic constraints: nonstrongly convex problems
- Cloud-Enabled Differentially Private Multiagent Optimization With Constraints
- Convergence of Asynchronous Distributed Gradient Methods Over Stochastic Networks
- Decentralized Optimization Over Time-Varying Directed Graphs With Row and Column-Stochastic Matrices
- Differentially Private Distributed Convex Optimization via Functional Perturbation
- Distributed Optimization Over Time-Varying Directed Graphs
- Distributed Spectrum Sensing for Cognitive Radio Networks by Exploiting Sparsity
- Distributed Subgradient Methods for Multi-Agent Optimization
- Dynamic average consensus under limited control authority and privacy requirements
- Enabling Privacy-Preservation in Decentralized Optimization
- EXTRA: an exact first-order algorithm for decentralized consensus optimization
- Harnessing Smoothness to Accelerate Distributed Optimization
- Notions of Centralized and Decentralized Opacity in Linear Systems
- On the convergence of decentralized gradient descent
- Privacy and safety analysis of timed stochastic discrete event systems using Markovian trajectory-observers
- Privacy Preserving Average Consensus
- Privacy preserving distributed optimization using homomorphic encryption
- Privacy-Preserving Distributed Averaging via Homomorphically Encrypted Ratio Consensus
- Privacy-Preserving Distributed Optimization via Subspace Perturbation: A General Framework
- Secure and Privacy-Preserving Consensus
- Verification of initial-state opacity in security applications of discrete event systems
Cited in
(13)- Protecting privacy through distributed computation in multi-agent decision making
- Design of Privacy-Preserving Dynamic Controllers
- Enabling Privacy-Preservation in Decentralized Optimization
- Decentralized Stochastic Optimization With Inherent Privacy Protection
- Quantization Enabled Privacy Protection in Decentralized Stochastic Optimization
- Decentralized nonconvex optimization with guaranteed privacy and accuracy
- Gradient-tracking based differentially private distributed optimization with enhanced optimization accuracy
- Robust privacy-preserving gossip averaging
- Distributed optimization for economic dispatch with acceleration and privacy preservation over unbalanced directed networks
- AB /Push-Pull method for distributed optimization in time-varying directed networks
- Privacy-preserving distributed adaptive estimation for non-stationary regression data
- Privacy-preserving gradient tracking via state decomposition for directed multi-agent optimization
- A weighted gradient tracking privacy-preserving method for distributed optimization
This page was built for publication: Dynamics based privacy preservation in decentralized optimization
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6163983)
