1: \begin{abstract}

2: 

3: We study the problem of network regression, where one is interested in how the topology of a network changes as a function of Euclidean covariates. We build upon recent developments in generalized regression models on metric spaces based on Fr\'echet means and propose a network regression method using the Wasserstein metric. We show that when representing graphs as multivariate Gaussian distributions, the network regression problem requires the computation of a Riemannian center of mass (i.e., Fr\'echet means). Fr\'echet means with non-negative weights translates into a barycenter problem and can be efficiently computed using fixed point iterations. Although the convergence guarantees of fixed-point iterations for the computation of Wasserstein affine averages remain an open problem, we provide evidence of convergence in a large number of synthetic and real-data scenarios.  

4: %The proposed approach applies Wasserstein barycenter method on Gaussian distributions with  to use the metric regression framework with the computational stability of barycenter methods and show empirically its superior performance.

5: Extensive numerical results show that the proposed approach improves existing procedures by accurately accounting for graph size, topology, and sparsity in synthetic experiments. Additionally, real-world experiments using the proposed approach result in higher Coefficient of Determination ($\bf{R^{2}}$) values and lower mean squared prediction error (MSPE), cementing improved prediction capabilities in practice. 

6: 

7: \end{abstract}

8: