Papers

Louis M. Shekhtman, Alexander J. Gates, Albert-László Barabási
While philanthropic support plays an increasing role in supporting research, there is limited quantitative knowledge about the patterns that characterize the distribution of philanthropic support. Here, we map philanthropic funding to universities and research institutions based on IRS tax forms from 685,397 non-profit organizations. We identify nearly one million grants supporting institutions involved in science, finding that in volume and scope, philanthropic funding is comparable to federal research funding. However, whereas federal funding relies on a few large organizations to distribute grants, the philanthropic ecosystem's support is fragmented among a large number of funders with diverse focus that support research institutions at varying levels. Furthermore, we find that distinct from government support, philanthropic funders tend to focus locally, indicating that other criteria, beyond research excellence, play a role in their funding decisions. We also show evidence of persistence, i.e., once a grant-giving relationship begins, it tends to continue in time. Finally, we discuss the policy implications of our findings for philanthropic funders, individual researchers, the science of science, and for quantitative studies of philanthropy in general.

Read the full article at: arxiv.org

Hao Peng, Daniel M. Romero, and Emőke-Ágnes Horvát

PNAS June 14, 2022 119 (25) e2119086119

Scientific retraction has been on the rise recently. Retracted papers are frequently discussed online, enabling the broad dissemination of potentially flawed findings. Our analysis spans a nearly 10-y period and reveals that most papers exhaust their attention by the time they get retracted, meaning that retractions cannot curb the online spread of problematic papers. This is striking as we also find that retracted papers are pervasive across mediums, receiving more attention after publication than nonretracted papers even on curated platforms, such as news outlets and knowledge repositories. Interestingly, discussions on social media express more criticism toward subsequently retracted results and may thus contain early signals related to unreliable work.

Read the full article at: www.pnas.org

Fengli Xu, Lingfei Wu, and James Evans

PNAS 119 (23) e2200927119

With teams growing in all areas of scientific and scholarly research, we explore the relationship between team structure and the character of knowledge they produce. Drawing on 89,575 self-reports of team member research activity underlying scientific publications, we show how individual activities cohere into broad roles of 1) leadership through the direction and presentation of research and 2) support through data collection, analysis, and discussion. The hidden hierarchy of a scientific team is characterized by its lead (or L) ratio of members playing leadership roles to total team size. The L ratio is validated through correlation with imputed contributions to the specific paper and to science as a whole, which we use to effectively extrapolate the L ratio for 16,397,750 papers where roles are not explicit. We find that, relative to flat, egalitarian teams, tall, hierarchical teams produce less novelty and more often develop existing ideas, increase productivity for those on top and decrease it for those beneath, and increase short-term citations but decrease long-term influence. These effects hold within person—the same person on the same-sized team produces science much more likely to disruptively innovate if they work on a flat, high-L-ratio team. These results suggest the critical role flat teams play for sustainable scientific advance and the training and advancement of scientists.

Read the full article at: www.pnas.org

Giulia Menichetti & Albert-László Barabási 

Nature Food volume 3, pages375–382 (2022)

Extensive programmes around the world endeavour to measure and catalogue the composition of food. Here we analyse the nutrient content of the full US food supply and show that the concentration of each nutrient follows a universal single-parameter scaling law that accurately captures the eight orders of magnitude in nutrient content variability. We show that the universality is rooted in the biochemical constraints obeyed by the metabolic pathways responsible for nutrient modulation, allowing us to confirm the empirically observed scaling law and to predict its variability in agreement with the data. We propose that the natural nutrient variability in food can be quantitatively formalized. This provides a mathematical rationale for imputing missing values in food composition databases and paves the way towards a quantitative understanding of the impact of food processing on nutrient balance and health effects.

Read the full article at: www.nature.com

Milo Z. Trujillo, Laurent Hébert-Dufresne & James Bagrow 

EPJ Data Science volume 11, Article number: 31 (2022)

GitHub has become the central online platform for much of open source, hosting most open source code repositories. With this popularity, the public digital traces of GitHub are now a valuable means to study teamwork and collaboration. In many ways, however, GitHub is a convenience sample, and may not be representative of open source development off the platform. Here we develop a novel, extensive sample of public open source project repositories outside of centralized platforms. We characterized these projects along a number of dimensions, and compare to a time-matched sample of corresponding GitHub projects. Our sample projects tend to have more collaborators, are maintained for longer periods, and tend to be more focused on academic and scientific problems.

Read the full article at: epjdatascience.springeropen.com

Somayeh Kalantari, Eslam Nazemi & Behrooz Masoumi
Computing (2022)

Nowadays, the study of self-organizing systems has attracted much attention. However, since these systems are run in dynamic, changing, and evolving environments, it is possible that undesirable behaviors that are contrary to the system goals occur. Therefore, it is necessary to provide mechanisms to guide the self-organizing system. However, several approaches were proposed to guide self-organizing systems, more effective approaches are required due to the variation of the contexts in which they are deployed and their complexity. This paper aims to use the self-advising property to provide guidelines about the context of self-organizing systems. The agents of these systems are guided implicitly by using the guidelines provided. In the proposed approach, contextual data is made by an advisor agent that produces them based on the agents’ behavioral entropy. The proposed approach is evaluated using a case study based on the NASA ANTS mission. According to experiments, the proposed approach causes adaptation activities’ costs to decrease at all radio ranges. Besides, in some radio ranges, i.e., 110 and 120 GHz, the guiding state’s adaptive time is less than the no-guiding state’s adaptive time. The evaluations also show that the ruler agents’ mean entropy in the guiding state is less than the no-guiding state in 75 % of radio ranges. This approach’s success in reducing the agents’ entropy indicates its ability to guide self-organizing systems.

Read the full article at: link.springer.com

Silja Sormunen, Thilo Gross, Jari Saramäki
It has been postulated that the brain operates in a self-organized critical state that brings multiple benefits, such as optimal sensitivity to input. Thus far, self-organized criticality has typically been depicted as a one-dimensional process, where one parameter is tuned to a critical value. However, the number of adjustable parameters in the brain is vast, and hence critical states can be expected to occupy a high-dimensional manifold inside a high-dimensional parameter space. Here, we show that adaptation rules inspired by homeostatic plasticity drive a neuro-inspired network to drift on a critical manifold, where the system is poised between inactivity and persistent activity. During the drift, global network parameters continue to change while the system remains at criticality.

Read the full article at: arxiv.org

Rafael Prieto Curiel, Humberto González Ramírez, and Steven Bishop

Front. Phys., 16 June 2022

A tragedy of the commons is said to occur when individuals act only in their own interest but, in so doing, create a collective state of a group that is less than optimal due to uncoordinated action. Here, we explore the individual decision-making processes of commuters using various forms of transport within a city, forming a modal share which is then built into a dynamical model using travel time as the key variable. From a randomised start in the distribution of the modal share, assuming that some individuals change their commuting method, favouring lower travel times, we show that a stable modal share is reached corresponding to an equilibrium in the model. Considering the average travel time for all commuters within the city, we show that an optimal result is achieved only if the direct and induced factors and the number of users are equal for all transport modes. For asymmetric factors, the equilibrium reached is always sub-optimal, leading to city travel trajectories being “tragic”, meaning that individuals choose a faster commuting time but create a slower urban mobility as a collective result. Hence, the city evolves, producing longer average commuting times. It is also shown that if a new mode of transport has a small baseline commuting time but has a high induced impact for other users, then introducing it might result in a counter-intuitive result producing more congestion, rather than less.

Read the full article at: www.frontiersin.org

Alexander F. Siegenfeld, Yaneer Bar-Yam
Ashby's law of requisite variety allows a comparison of systems with their environments, providing a necessary (but not sufficient) condition for system efficacy: a system must possess at least as much complexity as any set of environmental behaviors that require distinct responses from the system. However, the complexity of a system depends on the level of detail, or scale, at which it is described. Thus, the complexity of a system can be better characterized by a complexity profile (complexity as a function of scale) than by a single number. It would therefore be useful to have a multi-scale generalization of Ashby's law that requires that a system possess at least as much complexity as the relevant set of environmental behaviors *at each scale*. We construct a formalism for a class of complexity profiles that is the first, to our knowledge, to exhibit this multi-scale law of requisite variety. This formalism not only provides a characterization of multi-scale complexity but also generalizes the single constraint on system behaviors provided by Ashby's law to an entire class of multi-scale constraints. We show that these complexity profiles satisfy a sum rule, which reflects the important tradeoff between smaller- and larger-scale degrees of freedom, and we extend our results to subdivided systems and systems with a continuum of components.

Read the full article at: arxiv.org

José Balsa-Barreiro, Mónica Menendez & Alfredo J. Morales 

Scientific Reports volume 12, Article number: 9037 (2022)

The social space refers to physical or virtual places where people interact with one another. It decisively influences the emergence of human behaviors. However, little is known about the nature and complexity of the social space, nor its relationship to context and spatial scale. Recently, the science of complex systems has bridged between fields of knowledge to provide quantitative responses to fundamental sociological questions. In this paper, we analyze the shifting behavior of social space in terms of human interactions and wealth distribution across multiple scales using fine-grained data collected from both official (US Census Bureau) and unofficial data sources (social media). We use these data to unveil how patterns strongly depend upon the observation scale. Therefore, it is crucial for any analysis to be framed within the appropriate context to avoid biased results and/or misleading conclusions. Biased data analysis may lead to the adoption of fragile and poor decisions. Including context and a proper understanding of the spatial scale are essential nowadays, especially with the pervasive role of data-driven tools in decision-making processes.

Read the full article at: www.nature.com

Jiazhen Liu, Shengda Huang, Nathan Aden, Neil Johnson, Chaoming Song
Polarization is a ubiquitous phenomenon in social systems. Empirical studies show substantial evidence for opinion polarization across social media. Recent modeling works show qualitatively that polarization emerges in coevolving networks by integrating reinforcing mechanisms and network evolution. However, a quantitative and comprehensive theoretical framework capturing generic mechanisms governing polarization remains unaddressed. In this paper, we discover a universal scaling law for opinion distributions, characterized by a set of scaling exponents. These exponents classify social systems into polarization and depolarization phases. We find two generic mechanisms governing the polarization dynamics, and propose a coevolving framework that counts for opinion dynamics and network evolution simultaneously. We show analytically three different phases including polarization, partial polarization, and depolarization, and the corresponding phase diagram. In the polarized phase, our theory predicts that a bi-polarized community structure emerges naturally from the coevolving dynamics. These theoretical predictions are in line with observations in empirical datasets. Our theory not only accounts for the empirically observed scaling laws but also allows us to quantitatively predict scaling exponents.

Read the full article at: arxiv.org

Abeba Birhane, David J. T. Sumpter
When mathematical modelling is applied to capture a complex system, multiple models are often created that characterize different aspects of that system. Often, a model at one level will produce a prediction which is contradictory at another level but both models are accepted because they are both useful. Rather than aiming to build a single unified model of a complex system, the modeller acknowledges the infinity of ways of capturing the system of interest, while offering their own specific insight. We refer to this pragmatic applied approach to complex systems -- one which acknowledges that they are incompressible, dynamic, nonlinear, historical, contextual, and value-laden -- as Open Machine Learning (Open ML). In this paper we define Open ML and contrast it with some of the grand narratives of ML of two forms: 1) Closed ML, ML which emphasizes learning with minimal human input (e.g. Google's AlphaZero) and 2) Partially Open ML, ML which is used to parameterize existing models. To achieve this, we use theories of critical complexity to both evaluate these grand narratives and contrast them with the Open ML approach. Specifically, we deconstruct grand ML `theories' by identifying thirteen 'games' played in the ML community. These games lend false legitimacy to models, contribute to over-promise and hype about the capabilities of artificial intelligence, reduce wider participation in the subject, lead to models that exacerbate inequality and cause discrimination and ultimately stifle creativity in research. We argue that best practice in ML should be more consistent with critical complexity perspectives than with rationalist, grand narratives.

Read the full article at: arxiv.org