Kidz Time! Mattia Mattei: SYSMA+ NETWORKS=LOVE

Dear all,

I hope to find you well 🙂

We have a very special event: an (online) seminar of our young Mattia Mattei (TOFFEe project) today at 11.30. The link to the seminar is here: https://meet.google.com/ito-gwkx-byc

The title of his talk is An entropy-based perspective on Online Social Networks: semantic networks and bow-tie structures.
Abstract: My presentation will focus on two different applications of entropy-based models to the analysis of Online Social Networks, specifically Twitter. The first part of the meeting will be dedicated to the analysis of the Twitter semantic network during the Italian Covid-19 epidemic. Using as a benchmark an entropy-based bipartite configuration model (BiCM, [1]), we constructed a bipartite network of users and hashtags, related to the Italian online debate about the epidemic, following a strategy similar to the one developed in [2]. We observed that the debate is mainly political, even if the subject is not exclusively so. Remarkably, the observed discursive communities are not equally exposed to d/misinformation campaigns, as can be observed by considering the communities of hashtags related to false information about the origin of the COVID-19 pandemic, confirming previous findings [3].In the second work we characterise the network structure of the discursive communities. The bow-tie structure was initially introduced by Broder et al. [4] in order to study the World Wide Web, dividing, mainly, the entire system in a Strongly Connected Component (SCC) of all websites, in a IN block, including all search engines and an OUT block composed by the authorities (i.e. Wikipedia). Remarkably, the same structure is almost ubiquitous in the discursive communities on Twitter that focus on political subjects: we analyse 8 different discussions in different languages and a strong bow-tie structure is almost always present in many political discoursive communities, while it is absent in other different debates. We further characterize the various groups and blocks in the network. 

[1] Saracco et al, Scientific Reports (2016)
[2] Radicioni et al, arXiv:2009.02960 (2020)
[3] Caldarelli et al, arXiv:2010.01913 (2020)
[4] Broder et al, Computer Networks (2000) 

Be there and stay tuNedo!

Nedo

Advertisement

OnliNedo Seminar: Martin Rosvall!

Dear Nedoers,

I hope to find you well and safe!

The OnliNedo season 2020/2021 is going to finish with the last seminar: Martin Rosvall from Umeå University is going to give the seminar entitled Mapping flows on weighted and directed networks with incomplete observations. The talk is going to be held on Google Meet at the link https://meet.google.com/hys-dmpy-axh on Wednesday 12th of May at 14:30.

Abstract: Detecting significant community structure in networks with incomplete observations is challenging because the evidence for specific solutions fades away with missing data. For example, recent research shows that flow-based community detection methods can highlight spurious communities in sparse undirected and unweighted networks with missing links.

Missing link observations in weighted and directed networks aggravate this overfitting problem since each link carries more information.

To address this problem, here we extend the idea behind the Bayesian estimate of the map equation for unweighted and undirected networks to enable more robust community detection in weighted and directed networks. We derive a weighted and directed prior network that can incorporate metadata information and show that an efficient implementation in the community-detection method Infomap provides more reliable communities even with a significant fraction of the data missing.

Be there and stay tuNedo and safe!

Nedo

OnliNedo Seminars: Miguel A. Muñoz!

Dear Nedoers,

I hope to find you well and safe! This special April continues with prof. Miguel A. Muñoz, University of Grenada!

PAY ATTENTION!
The date and time changed: the seminar will be held next Thursday 15th April at 12.00. The link to Google Meet is: https://meet.google.com/cws-ctgm-oac

The title of the seminar is Rapid evolution of diversified survival strategies in bacterial communities.

Abstract: Ecological and evolutionary dynamics have been historically regarded as unfolding at broadly separated timescales.  However, these two types of processes are nowadays well-documented to intermingle much more tightly than traditionally assumed, especially in communities of microorganisms. Improving the existing approches to these problems and developing novel analytical and computational frameworks is a challenge of utmost theoretical and practical relevance.

Within this context, in this talk I present an introduction to some relevant concepts such as “bet-hedging”, optimal strategies, evolutionary tradoffs, diversification, etc. Then, I discuss recent experimental results showing empirical evidence of very rapid evolution of tolerance by lag in bacterial communities that are periodically exposed to antibiotic stress in laboratory conditions.  In particular, the communities evolve to develop a broad distribution of lag times —i.e. the times that individual bacteria typically remain in a dormant state to cope with stress— whose statistics are determined by those of the environments, thus creating a sort of collective memory in the community. Finally, I will discuss a general analytical and computational framework that we have developed to model and shed light onto this specific example as well as on more general eco-evolutionary problems and discuss further directions and open problems.

Stay safe and tuNedo,

cheers,

Nedo

OnliNedo seminars: Subodh P. Patil!

Dear Nedoers,

I hope to find you well 🙂 April is going to be full of new seminars: next Friday 9th we are going to have Chiara Poletto, on the 16th we are going to have Miguel Muñoz (more details to come) and on the 22nd we are going to have Subodh P. Patil from the Instituut-Lorentz for Theoretical Physics of Leiden with the seminar “Random Networks and Geometries in Physics“.

Abstract: Physicists have long speculated that spacetime at the very shortest scales may not be a continuum. I will attempt to provide an accessible overview of various attempt to construct spacetime geometries from first principles that draw upon, or are adjacent to ideas in network theory. This talk assumes no prior knowledge beyond the basics of network theory (although some knowledge of the basics of information theory will help for context), and will culminate in a review of the recent preprint “The birth of geometry in exponential random graphs” https://arxiv.org/abs/2102.11477

The seminar is going to be on Google Meet at https://meet.google.com/cco-gjia-nzw.

Stay tuNedo and safe,
cheers,

Nedo

Nedo Award: Online Champion!

Dear Nedoers,

we all saw multiple effects of the pandemic on our daily life. For researchers, it means that every workshop, conference or seminar cannot be held in person. So, we expected that the Nedo Award would freeze for a while.
We could not be more wrong: at the end of his presentation at the APS March Meeting, Tommaso Gili ended his presentation with the following slide:

Congratulations to Tommaso, the new winner of the Nedo Award!

Stay tuNedo and safe,

Nedo

OnliNedo Seminar: Chiara Poletto!

Dear Nedoers,

I hope to find you well and safe 🙂

April is coming and the next OnliNedo seminar is approaching: next 9h of April at 12 we have Chiara Poletto, researcher at the  Pierre Louis Institute of Epidemiology and Public Health, as part of the INSERM (France). She is going to present the seminar Diversity in humans and pathogens: effect on epidemics and the impact of interventions on Google Meet at the following address: https://meet.google.com/ini-ciwx-ywq

Abstract: In the last decades, new network theories and epidemiological evidences have substantially advanced our knowledge about how acute respiratory infections spread in the human population. Temporal and multi-layer networks provide a paradigmatic example. These frameworks allow for clearly describing the heterogeneous connectivity features driving transmission, e.g. different connectivity by settings (household, workplace, school, etc.), and occasional vs. recurrent encounters. However, intrinsic properties of individuals and pathogens, besides individuals’ connectivity are central in shaping the dynamics of transmission. For instance, individuals may be heterogeneous in their susceptibility and their probability to develop a severe form of infection. On the other hand, heterogeneous pathogens and pathogen strains alter the dynamics and burden of an epidemic and are among the drivers of pathogen emergence events. These levels of complexity mutually affect each other which makes it difficult to disentangle their relative role on the epidemic unfolding and the effect of interventions. During the talk I will tackle this issue, presenting recent theoretical and applied works. I will review some studies on the COVID-19 pandemic and discuss the implications of multi-strain/multi-pathogen interaction for the emergence of new pathogens and their co-existence.

Stay safe and be tuNedo,
cheers!

Nedo

OnliNedo Seminar: Renaud Lambiotte!

Dear Nedoers,

I hope to find you well and safe 🙂 The March OnliNedo Seminar is “Higher-order networks and Nonlinear consensus” by prof. Renaud Lambiotte (University of Namur; Oxford University), next 11th of March at 12:00 on Google Meet at https://meet.google.com/prz-dnam-drz

Abstract: Over the last 20 years, networks have emerged as a powerful paradigm to model complex systems. Despite its many successes, the abstraction of a system in terms of nodes and edges has also some fundamental modelling limitations, that have become more apparent with the increasing availability of multi-way relational data recently. Within the emerging field of higher-order networks, researchers have tested the limits of the network paradigm, and proposed extensions with richer interactions, including multiplex networks, higher-order Markov models and multiway networks. In this talk, I will start by providing an overview of recent results on higher-order networks, before turning to multiway networks and focusing on the problem of non-linear consensus on hypergraphs.

Be there and stay tuNedo!
Cheers,

Nedo

p.s. Thanks to Renaud, the slides are available here.

OnliNedo Seminars special February: Norbert Marwan and Michel Mandjes!

Dear Nedoers,

I hope to find you well and safe 🙂

Next February is going to be quite filled of OnliNedo seminars!
We start on the 4th of February at 14 with prof. Norbert Marwan (Potsdam Institute for Climate Impact Research) with the seminar “Reconstructing complex networks from data“.
Abstract: Complex networks provide an interesting tool to investigate spatio-temporal data. The first step is to reconstruct a (functional) network from data. I will show different reconstruction approaches depending on the research question and the nature of the data. The procedure is illustrated with applications on climate data.
Pay attention! Due to some technical issues, the seminar will not be on Google Meet, but on Zoom at the address:  https://pik-potsdam.zoom.us/j/92187003129?pwd=QTBuU1FkNWg5dG9lV2VmQkFBbm5OQT09

Then, the next 10th of February, at 12 we have prof. Michel Mandjes (VU University Amsterdam) with the seminar: “A Diffusion-based analysis of a road traffic network“.
Abstract: In this talk I will discuss an important example of complex networks, namely road traffic networks. I start by giving an overview of the existing literature, distinguishing between microscopic models (describing the stochastic evolution of the position of individual vehicles) and macroscopic models (built around deterministic continuous flows, represented as partial differential equations). Then I argue that the “optimal” model is a compromise between these: we aim at a stochastic model with enough aggregation to make sure that explicit limiting analysis can be performed. The underlying dynamics are consistent with the macroscopic fundamental diagram that describes the functional relation between the vehicle density and velocity. Discretizing space, the model can be phrased in terms of a spatial population process, thus allowing the application of a classical scaling approach. More specifically, it follows that under a diffusion scaling, the vehicle density process can be approximated by an appropriate Gaussian process. This Gaussian approximation can be used to evaluate the travel time distribution between a given origin and destination.
In this case, we go back to GMeet at https://meet.google.com/hdv-zcxa-zvw

Stay tuNedo and safe and do not miss the Special February onliNedo seminars 🙂

Nedo

OnliNedo Seminars: Naoki Masuda!

Dear Nedoers,

Happy Ne(do)w Year 🙂 Even if OnliNedo seminars took a Xmas vacation on December, the program of this year (available on the left column of the website) is super cool and dense!

We are starting the new year with the OnliNedo Seminar by Naoki Masuda, Associate Professor at Department of Mathematics, University at Buffalo. The title of the seminar is A two-state view of bursts and heavy tails in contact networks and it is going to take place on Google Meet at https://meet.google.com/ksq-uwqg-zvf on Friday 22 January at 16 (CET).

Abstract: Many aspects of human (and animal) activity, such as the frequency of contacts of an individual, the number of interaction partners, and the time between the contacts of two individuals, are characterized by heavy-tailed distributions. Heavy-tailed nature of interevent times has a large impact on dynamical processes occurring on networks and populations, such as contagion. In this presentation, we introduce a two-state modeling approach in which each node switches between a high-activity and low-activity states over time in a Markovian manner. This assumption facilitates theoretical analyses and also provides models that mimic heavy-tailed distributions of interevent times because a mixture of two (or a few) exponential distributions phenomenologically look similar to heavy-tailed distributions across a practically relevant scale. In the first part of the presentation, we present a model of interevent times in which each pair of individuals interacts at an enhanced rate if and only if both of them are in the high-activity state (i.e., they chat with each other only when both of them want to). In the second part, we analyze the susceptible-infected-susceptible/recovered (SIS/R) models when each node similarly switches between two states over time. We analytically and numerically argue that it is not the tail but the small values of interevent times that control spreading dynamics.

Be there and stay tuNedo and safe,

Nedo

OnliNedo seminar: Bernardo Monechi!

Dear Nedoers,

As announced, Nedo strikes back and no pandemic will stop it!
So, from now on, every second Friday of the month, we will have a new OnliNedo seminar. The next speaker is our friend and colleague Bernardo Monechi, from SONY CSL, Paris.

The OnliNedo seminar is going to be on Google Meet at the following link: meet.google.com/rdh-shfb-tjh on Friday 13 November at 14:30. The title of the seminar is “Maximum Entropy Approach to Macro-Economic Urban Dynamics“.

Abstract: The rapid urbanization makes the understanding of the evolution of urban environments of utmost importance to steer societies towards better futures. Many studies have focused on the emerging properties of cities, leading to the discovery of scaling laws mirroring the dependence of socio-economic indicators on city sizes. However, few efforts have been devoted to the modelling of the dynamical evolution of cities, as reflected through the mutual influence of socio-economic variables. Here, we fill this gap by presenting a maximum entropy generative model for cities written in terms of a few macro-economic variables, whose parameters (the effective Hamiltonian, in a statistical-physical analogy) are inferred from real data through a maximum-likelihood approach. This approach allows for establishing a few results. First, nonlinear dependencies among indicators are needed for an accurate statistical description of the complexity of empirical correlations. Second, the inferred coupling parameters turn out to be quite robust along different years. Third, the quasi time-invariance of the effective Hamiltonian allows guessing the future state of a city based on a previous state.

Be there and stay tuNedo and safe,

Nedo

Among other things, Bernardo is a professional Lego player.