1: \begin{abstract}

2:     This paper proposes a novel class of distributed continuous-time

3:     coordination algorithms to solve network optimization problems

4:     whose cost function is a sum of local cost functions associated to

5:     the individual agents.  We establish the exponential convergence

6:     of the proposed algorithm under (i) strongly connected and

7:     weight-balanced digraph topologies when the local costs are

8:     strongly convex with globally Lipschitz gradients, and (ii)

9:     connected graph topologies when the local costs are strongly

10:     convex with locally Lipschitz gradients.  When the local cost

11:     functions are convex and the global cost function is strictly

12:     convex, we establish asymptotic convergence under connected graph

13:     topologies.  We also characterize the algorithm's correctness

14:     under time-varying interaction topologies and study its privacy

15:     preservation properties. Motivated by practical considerations, we

16:     analyze the algorithm implementation with discrete-time

17:     communication.  We provide an upper bound on the stepsize that

18:     guarantees exponential convergence over connected graphs for

19:     implementations with periodic communication.  Building on this

20:     result, we design a provably-correct centralized event-triggered

21:     communication scheme that is free of Zeno behavior. Finally, we

22:     develop a distributed, asynchronous event-triggered communication

23:     scheme that is also free of Zeno with asymptotic convergence

24:     guarantees.  Several simulations illustrate our results.

25:   \end{abstract}

26: