Preserving Statistical Privacy in Distributed Optimization
This addresses privacy concerns in distributed systems for applications like multi-agent learning, though it is incremental as it builds on existing non-private methods.
The paper tackles the problem of preserving statistical privacy in distributed optimization by proposing a protocol that protects agents' local cost functions against a passive adversary corrupting up to t agents, ensuring accuracy through a zero-sum obfuscation method that maintains the sum of cost functions.
We present a distributed optimization protocol that preserves statistical privacy of agents' local cost functions against a passive adversary that corrupts some agents in the network. The protocol is a composition of a distributed ``{\em zero-sum}" obfuscation protocol that obfuscates the agents' local cost functions, and a standard non-private distributed optimization method. We show that our protocol protects the statistical privacy of the agents' local cost functions against a passive adversary that corrupts up to $t$ arbitrary agents as long as the communication network has $(t+1)$-vertex connectivity. The ``{\em zero-sum}" obfuscation protocol preserves the sum of the agents' local cost functions and therefore ensures accuracy of the computed solution.