1 paper
Alexander Rogozin, Mikhail Bochko, Pavel Dvurechensky +2
We consider a distributed stochastic optimization problem that is solved by a decentralized network of agents with only local communication between neighboring agents. The goal of…