Dynamics based privacy preservation in decentralized optimization

With decentralized optimization having increased applications in various domains ranging from machine learning, control, to robotics, its privacy is also receiving attention. Existing solutions for achieve by patching information-technology mechanisms such as differential or homomorphic encryption, which either sacrifices accuracy incurs heavy computation/communication overhead. We propose an inherently privacy-preserving algorithm exploiting the robustness of dynamics. More specifically, we present a general framework, based on show that participating nodes’ gradients can be protected adding randomness parameters. further added has no influence optimization, and prove our R-linear convergence when global objective function smooth strongly convex. proposed avoid gradient node being inferable other nodes. Simulation results confirm theoretical predictions.