Privacy-Preserving Distributed Optimization Algorithm With General Local Constraints on Weight-Unbalanced Digraph

In this work, a differential privacy discrete-time distributed optimization algorithm is proposed in order to solve the convex optimization problems on unbalanced directed graph while accommodating general local constraints. With the privacy of all participating nodes ensured, the proposed distributed algorithm can minimize the sum of local objective functions, which is particularly important in situations where an external eavesdropper may intercept node interactions. Furthermore, this paper demonstrates that there is a trade-off between random noise and the convergence rate of the algorithm. Finally, simulation experiments validate the findings of this paper.