  


Distributed Gradient Methods with Variable Number of Working Nodes
Dusan Jakovetic(djakovetuns.ac.rs) Abstract: We consider distributed optimization where $N$ nodes in a connected network minimize the sum of their local costs subject to a common constraint set. We propose a distributed projected gradient method where each node, at each iteration $k$, performs an update (is active) with probability $p_k$, and stays idle (is inactive) with probability $1p_k$. Whenever active, each node performs an update by weightaveraging its solution estimate with the estimates of its active neighbors, taking a negative gradient step with respect to its local cost, and performing a projection onto the constraint set; inactive nodes perform no updates. Assuming that nodes' local costs are strongly convex, with Lipschitz continuous gradients, we show that, as long as activation probability $p_k$ grows to one asymptotically, our algorithm converges in the mean square sense (MSS) to the same solution as the standard distributed gradient method, i.e., as if all the nodes were active at all iterations. Moreover, when $p_k$ grows to one linearly, with an appropriately set convergence factor, the algorithm has a linear MSS convergence, with practically the same factor as the standard distributed gradient method. Simulations demonstrate that, when compared with the standard distributed gradient method, the proposed algorithm significantly reduces the overall number of pernode communications and pernode gradient evaluations (computational cost) for the same required accuracy. Keywords: Distributed optimization, distributed gradient method, variable number of working nodes, convergence rate, consensus. Category 1: Network Optimization Category 2: Convex and Nonsmooth Optimization (Convex Optimization ) Citation: Download: [PDF] Entry Submitted: 04/15/2015 Modify/Update this entry  
Visitors  Authors  More about us  Links  
Subscribe, Unsubscribe Digest Archive Search, Browse the Repository

Submit Update Policies 
Coordinator's Board Classification Scheme Credits Give us feedback 
Optimization Journals, Sites, Societies  