Segregation dynamics with reinforcement learning and agent based modeling

Download

10.1038:s41598-020-68447-8.pdf

Date

2020-7-16

Author

Sert, Egemen
Bar-Yam, Yaneer
Morales, Alfredo J.

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

172
views

131
downloads

Societies are complex. Properties of social systems can be explained by the interplay and weaving of individual actions. Rewards are key to understand people's choices and decisions. For instance, individual preferences of where to live may lead to the emergence of social segregation. In this paper, we combine Reinforcement Learning (RL) with Agent Based Modeling (ABM) in order to address the self-organizing dynamics of social segregation and explore the space of possibilities that emerge from considering different types of rewards. Our model promotes the creation of interdependencies and interactions among multiple agents of two different kinds that segregate from each other. For this purpose, agents use Deep Q-Networks to make decisions inspired on the rules of the Schelling Segregation model and rewards for interactions. Despite the segregation reward, our experiments show that spatial integration can be achieved by establishing interdependencies among agents of different kinds. They also reveal that segregated areas are more probable to host older people than diverse areas, which attract younger ones. Through this work, we show that the combination of RL and ABM can create an artificial environment for policy makers to observe potential and existing behaviors associated to rules of interactions and rewards.

Subject Keywords

Multidisciplinary

URI

https://hdl.handle.net/11511/51441

Journal

Scientific Reports

DOI

https://doi.org/10.1038/s41598-020-68447-8

Collections

Department of Electrical and Electronics Engineering, Article

Suggestions

OpenMETU
Core

Novel insights on demographic history of tribal and caste groups from West Maharashtra (India) using genome-wide data Debortoli, Guilherme; Abbatangelo, Cristina; Ceballos, Francisco; Fortes-Lima, Cesar; Norton, Heather L.; Ozarkar, Shantanu; Parra, Esteban J.; Jonnalagadda, Manjari (Springer Science and Business Media LLC, 2020-6-22) The South Asian subcontinent is characterized by a complex history of human migrations and population interactions. In this study, we used genome-wide data to provide novel insights on the demographic history and population relationships of six Indo-European populations from the Indian State of West Maharashtra. The samples correspond to two castes (Deshastha Brahmins and Kunbi Marathas) and four tribal groups (Kokana, Warli, Bhil and Pawara). We show that tribal groups have had much smaller effective popul...
Hamilton-Jacobi theory of discrete, regular constrained systems Güler, Y. (Springer Science and Business Media LLC, 1987-8) The Hamilton-Jacobi differential equation of a discrete system with constraint equationsG α=0 is constructed making use of Carathéodory’s equivalent Lagrangian method. Introduction of Lagrange’s multipliersλ˙α as generalized velocities enables us to treat the constraint functionsG α as the generalized momenta conjugate toλ˙α. Canonical equations of motion are determined.
Mechanical Fracturing of Core-Shell Undercooled Metal Particles for Heat-Free Soldering Çınar, Simge; Chen, Jiahao; Thuo, Martin (Springer Science and Business Media LLC, 2016-02-23) Phase-change materials, such as meta-stable undercooled (supercooled) liquids, have been widely recognized as a suitable route for complex fabrication and engineering. Despite comprehensive studies on the undercooling phenomenon, little progress has been made in the use of undercooled metals, primarily due to low yields and poor stability. This paper reports the use of an extension of droplet emulsion technique (SLICE) to produce undercooled core-shell particles of structure; metal/oxide shell-acetate ('/'=...
PREMARITAL BREAKUPS IN A TURKISH SAMPLE - PERCEIVED REASONS, ATTRIBUTIONAL DIMENSIONS AND AFFECTIVE REACTIONS HORTACSU, N; Karancı, Ayşe Nuray (1987-02-01) The present study aimed to investigate the causal attributional dimensions used in the dissolution of premarital romantic relationships by using the subjects' own ratings, and to examine cultural influences on such dimensions. Twenty‐two items tapping dimensions revealed by previous research were prepared and administered to 135 Turkish students who have broken up a relationship within the past year. Respondents' ratings of their perceived principal reason for dissolution were subjected to factor analysis w...
Decision making under uncertainty: factors influencing mate selection in a dating application simulation Semchenko, Ayten Yeşim; Çakır, Murat Perit; Temürcü, Ceyhan; Department of Cognitive Sciences (2017) Human social intentions are heavily influenced by their perceptions of others. In this study, the effect of perceived attractiveness is of particular interest. The purpose of this study is to explore the effect of perceived attractiveness on the likelihood of wanting to have sex and the likelihood of using condom among dating application users. In real dating application environments, there would be many uncontrolled variables (e.g., age, not/having common friends etc.) related to the profiles that the part...

Citation Formats

E. Sert, Y. Bar-Yam, and A. J. Morales, “Segregation dynamics with reinforcement learning and agent based modeling,” Scientific Reports, 2020, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/51441.