Dynamic networks are used in a wide range of fields, including social network analysis, recommender systems and epidemiology. Representing complex networks as structures changing over time allow network models to leverage not only structural but also temporal patterns. However, as dynamic network literature stems from diverse fields and makes use of inconsistent terminology, it is challenging to navigate. Meanwhile, graph neural networks (GNNs) have gained a lot of attention in recent years for their ability to perform well on a range of network science tasks, such as link prediction and node classification. Despite the popularity of graph neural networks and the proven benefits of dynamic network models, there has been little focus on graph neural networks for dynamic networks. To address the challenges resulting from the fact that this research crosses diverse fields as well as to survey dynamic graph neural networks, this work is split into two main parts. First, to address the ambiguity of the dynamic network terminology we establish a foundation of dynamic networks with consistent, detailed terminology and notation. Second, we present a comprehensive survey of dynamic graph neural network models using the proposed terminology.

Foundations and Modeling of Dynamic Networks Using Dynamic Graph Neural Networks: A Survey

Revisiting the complex adaptive systems paradigm: Leading perspectives for researching operations and supply chain management issues

Genetic time-series data from historical samples greatly facilitate inference of past population dynamics and species evolution. Yet, although climate and landscape change are often touted as post-hoc explanations of biological change, our understanding of past climate and landscape change influences on evolutionary processes is severely hindered by the limited application of methods that directly relate environmental change to species dynamics through time. Increased integration of spatiotemporal environmental and genetic data will revolutionize the interpretation of environmental influences on past population processes and the quantification of recent anthropogenic impacts on species, and vastly improve prediction of species responses under future climate change scenarios, yielding widespread revelations across evolutionary biology, landscape ecology and conservation genetics. This review encourages greater use of spatiotemporal landscape genetic analyses that explicitly link landscape, climate and genetic data through time by providing an overview of analytical approaches for integrating historical genetic and environmental data in five key research areas: population genetic structure, demography, phylogeography, metapopulation connectivity and adaptation. We also include a tabular summary of key methodological information, suggest approaches for mitigating the particular difficulties in applying these techniques to ancient DNA and palaeoclimate data, and highlight areas for future methodological development.

Spatiotemporal landscape genetics: Investigating ecology and evolution through space and time.

A dynamic network model with persistent links and node-specific latent variables, with an application to the interbank market

Given the growing number of available tools for modeling dynamic networks, the choice of a suitable model becomes central. The goal of this survey is to provide an overview of tie‐oriented dynamic network models. The survey is focused on introducing binary network models with their corresponding assumptions, advantages, and shortfalls. The models are divided according to generating processes, operating in discrete and continuous time. First, we introduce the temporal exponential random graph model (TERGM) and the separable TERGM (STERGM), both being time‐discrete models. These models are then contrasted with continuous process models, focusing on the relational event model (REM). We additionally show how the REM can handle time‐clustered observations, that is, continuous‐time data observed at discrete time points. Besides the discussion of theoretical properties and fitting procedures, we specifically focus on the application of the models on two networks that represent international arms transfers and email exchange, respectively. The data allow to demonstrate the applicability and interpretation of the network models.

https://onlinelibrary.wiley.com/doi/epdf/10.1111/stan.12198

Tempus volat, hora fugit: A survey of tie‐oriented dynamic network models in discrete and continuous time

We present a selective review of statistical modeling of dynamic networks. We focus on models with latent variables, specifically, the latent space models and the latent class models (or stochastic blockmodels), which investigate both the observed features and the unobserved structure of networks. We begin with an overview of the static models, and then we introduce the dynamic extensions. For each dynamic model, we also discuss its applications that have been studied in the literature, with the data source listed in Appendix. Based on the review, we summarize a list of open problems and challenges in dynamic network modeling with latent variables.

/pdf/a-review-of-dynamic-network-models-with-latent-variables-2fb33a6t03.pdf

A review of dynamic network models with latent variables

A Review of Dynamic Network Models with Latent Variables

For nearly eighty years, the National Geospatial-Intelligence Agency (NGA) has done its mapping—everything from the surface of the moon to the compound hiding Osama bin Laden—from the city of St. Louis. In 2017, this federal agency pitted three regional sites against each other in the competition to host their new billion-dollar headquarters. Two years later, a site of nearly one hundred acres of “underdeveloped brownfields” in North St. Louis was selected to go from “eyesore to economic boon” as the host of this moated and militarized spy headquarters. 1 This supposedly catalytic development eradicated nearly half of what was left of the St. Louis Place neighborhood following decades of erasure from urban renewal, malignant neglect, and unethical real estate transactions. While the project was touted as a massive win for St. Louis due primarily to preserving city income tax dollars, this essay examines how existing measures of success that conflate economic growth (or even simply economic preservation) with progress fail to capture the real social, environmental, and economic costs that development can bring. It considers how the difficult-to-measure aspects of city sociality—rootedness, optimism, opportunity, and social resilience, for example—are critical in recognizing and elevating the importance of spatial justice in our design decisions and design pedagogy. This paper explores the true costs of the NGA relocation to North St. Louis by measuring a broad range of social and economic factors ignored by the city’s assessment. It also considers a second adjacent project, the proposed Brickline Greenway, and explores how its more inclusive objectives begin taking small steps towards repair. Speculative design work from the Master of Urban Design studio at Washington University in St. Louis provides alternative mapping and design solutions that further this paradigm shift.

Bomin Kim

Papers

A review of dynamic network models with latent variables

A Review of Dynamic Network Models with Latent Variables

Measuring What Matters