Wilbert Samuel Rossi

Giacomo Como, Wilbert Samuel Rossi, Fabio Fagnani

The spread of new beliefs, behaviors, conventions, norms, and technologies in social and economic networks are often driven by cascading mechanisms, and so are contagion dynamics in financial networks. Global behaviors generally emerge from the interplay between the structure of the interconnection topology and the local agents' interactions. We focus on the Linear Threshold Model (LTM) of cascades first introduced by Granovetter (1978). This can be interpreted as the best response dynamics in a network game whereby agents choose strategically between two actions and their payoff is an increasing function of the number of their neighbors choosing the same action. Each agent is equipped with an individual threshold representing the number of her neighbors who must have adopted a certain action for that to become the agent's best response. We analyze the LTM dynamics on large-scale networks with heterogeneous agents. Through a local mean-field approach, we obtain a nonlinear, one-dimensional, recursive equation that approximates the evolution of the LTM dynamics on most of the networks of a given size and distribution of degrees and thresholds. Specifically, we prove that, on all but a fraction of networks with given degree and threshold statistics that is vanishing as the network size grows large, the actual fraction of adopters of a given action in the LTM dynamics is arbitrarily close to the output of the aforementioned recursion. We then analyze the dynamic behavior of this recursion and its bifurcations from a dynamical systems viewpoint. Applications of our findings to some real network testbeds show good adherence of the theoretical predictions to numerical simulations.

Wilbert Samuel Rossi, Paolo Frasca

This paper is about a new model of opinion dynamics with opinion-dependent connectivity. We assume that agents update their opinions asynchronously and that each agent's new opinion depends on the opinions of the $k$ agents that are closest to it. We show that the resulting dynamics is substantially different from comparable models in the literature, such as bounded-confidence models. We study the equilibria of the dynamics, observing that they are robust to perturbations caused by the introduction of new agents. We also prove that if the number of agents $n$ is smaller than $2k$, the dynamics converge to consensus. This condition is only sufficient.

Wilbert Samuel Rossi, Paolo Frasca, Fabio Fagnani

Important applications in robotic and sensor networks require distributed algorithms to solve the so-called relative localization problem: a node-indexed vector has to be reconstructed from measurements of differences between neighbor nodes. In a recent note, we have studied the estimation error of a popular gradient descent algorithm showing that the mean square error has a minimum at a finite time, after which the performance worsens. This paper proposes a suitable modification of this algorithm incorporating more realistic "a priori" information on the position. The new algorithm presents a performance monotonically decreasing to the optimal one. Furthermore, we show that the optimal performance is approximated, up to a 1 + \eps factor, within a time which is independent of the graph and of the number of nodes. This convergence time is very much related to the minimum exhibited by the previous algorithm and both lead to the following conclusion: in the presence of noisy data, cooperation is only useful till a certain limit.

Wilbert Samuel Rossi, Paolo Frasca

In this paper we elaborate upon a measure of node influence in social networks, which was recently proposed by Vassio et al., IEEE Trans. Control Netw. Syst., 2014. This measure quantifies the ability of the node to sway the average opinion of the network. Following the approach by Vassio et al., we describe and study a distributed message passing algorithm that aims to compute the nodes' influence. The algorithm is inspired by an analogy between potentials in electrical networks and opinions in social networks. If the graph is a tree, then the algorithm computes the nodes' influence in a number of steps equal to the diameter of the graph. On general graphs, the algorithm converges asymptotically to a meaningful approximation of the nodes' influence. In this paper we detail the proof of convergence, which greatly extends previous results in the literature, and we provide simulations that illustrate the usefulness of the returned approximation.

Wilbert Samuel Rossi, Paolo Frasca

The harmonic influence is a measure of the importance of nodes in social networks, which can be approximately computed by a distributed message-passing algorithm. In this extended abstract we look at two open questions about this algorithm. How does it perform on real social networks, which have complex topologies structured in communities? How does it perform when the network topology changes while the algorithm is running? We answer these two questions by numerical experiments on a Facebook ego network and on synthetic networks, respectively. We find out that communities can introduce artefacts in the final approximation and cause the algorithm to overestimate the importance of "local leaders" within communities. We also observe that the algorithm is able to adapt smoothly to changes in the topology.

Wilbert Samuel Rossi, Paolo Frasca, Fabio Fagnani

In our recent paper W.S. Rossi, P. Frasca and F. Fagnani, "Average resistance of toroidal graphs", SIAM Journal on Control and Optimization, 53(4):2541--2557, 2015, we studied how the average resistances of $d$-dimensional toroidal grids depend on the graph topology and on the dimension of the graph. Our results were based on the connection between resistance and Laplacian eigenvalues. In this note, we contextualize our work in the body of literature about random walks on graphs. Indeed, the average effective resistance of the $d$-dimensional toroidal grid is proportional to the mean hitting time of the simple random walk on that grid. If $d\geq3 $, then the average resistance can be bounded uniformly in the number of nodes and its value is of order $1/d$ for large $d$.

Wilbert Samuel Rossi, Jan Willem Polderman, Paolo Frasca

In online platforms, recommender systems are responsible for directing users to relevant contents. In order to enhance the users' engagement, recommender systems adapt their output to the reactions of the users, who are in turn affected by the recommended contents. In this work, we study a tractable analytical model of a user that interacts with an online news aggregator, with the purpose of making explicit the feedback loop between the evolution of the user's opinion and the personalised recommendation of contents. More specifically, we assume that the user is endowed with a scalar opinion about a certain issue and seeks news about it on a news aggregator: this opinion is influenced by all received news, which are characterized by a binary position on the issue at hand. The user is affected by a confirmation bias, that is, a preference for news that confirm her current opinion. The news aggregator recommends items with the goal of maximizing the number of user's clicks (as a measure of her engagement): in order to fulfil its goal, the recommender has to compromise between exploring the user's preferences and exploiting what it has learned so far. After defining suitable metrics for the effectiveness of the recommender systems (such as the click-through rate) and for its impact on the opinion, we perform both extensive numerical simulations and a mathematical analysis of the model. We find that personalised recommendations markedly affect the evolution of opinions and favor the emergence of more extreme ones: the intensity of these effects is inherently related to the effectiveness of the recommender. We also show that by tuning the amount of randomness in the recommendation algorithm, one can seek a balance between the effectiveness of the recommendation system and its impact on the opinions.

Wilbert Samuel Rossi, Paolo Frasca, Fabio Fagnani

The average effective resistance of a graph is a relevant performance index in many applications, including distributed estimation and control of network systems. In this paper, we study how the average resistance depends on the graph topology and specifically on the dimension of the graph. We concentrate on $d$-dimensional toroidal grids and we exploit the connection between resistance and Laplacian eigenvalues. Our analysis provides tight estimates of the average resistance, which are key to study its asymptotic behavior when the number of nodes grows to infinity. In dimension two, the average resistance diverges: in this case, we are able to capture its rate of growth when the sides of the grid grow at different rates. In higher dimensions, the average resistance is bounded uniformly in the number of nodes: in this case, we conjecture that its value is of order $1/d$ for large $d$. We prove this fact for hypercubes and when the side lengths go to infinity.