1: \begin{abstract}

2: There is a wealth of applied problems that can be posed as a dynamical

3: system defined on a network with both attractive and repulsive

4: interactions. Some examples include: understanding synchronization

5: properties of nonlinear oscillator;, the behavior of groups, or

6: cliques, in social networks; the study of optimal convergence for

7: consensus algorithm; and many other examples.  Frequently the

8: problems involve computing the index of a matrix,~i.e. the number of

9: positive and negative eigenvalues, and the dimension of the kernel. In

10: this paper we consider one of the most common examples, where the

11: matrix takes the form of a signed graph Laplacian. We show that the

12: there are topological constraints on the index of the Laplacian matrix

13: related to the dimension of a certain homology group. In certain

14: situations, when the homology group is trivial, the index of the

15: operator is rigid and is determined only by the topology of the

16: network and is independent of the strengths of the interactions. In

17: general these constraints give upper and lower bounds on the number of

18: positive and negative eigenvalues, with the dimension of the homology

19: group counting the number of eigenvalue crossings. The homology group

20: also gives a natural decomposition of the dynamics into ``fixed''

21: degrees of freedom, whose index does not depend on the edge-weights,

22: and an orthogonal set of ``free'' degrees of freedom, whose index

23: changes as the edge weights change. We also present some numerical

24: studies of this problem for large random matrices.

25: 

26: \end{abstract}

27: