Collective problem-solving in evolving networks: An agent-based model

Mohsen Jafari Songhori, César García-Díaz

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Research works in collective problem-solving usually assume fixed communication structures and explore effects thereof. In contrast, in real settings, individuals may modify their set of connections in the search of information and feasible solutions. This paper illustrates how groups collectively search for solutions in a space under the presence of dynamic structures and individual-level learning. For that, we built an agent-based computational model. In our model, individuals (i) simultaneously conduct search of solutions over a complex space (i.e. a NK landscape), (ii) are initially connected to each other according to a given network configuration, (iii) are endowed with learning capabilities (through a reinforcement learning algorithm), and (iv) update (i.e. create or severe) their links to other agents according to such learning features. Results reveal conditions under which performance differences are obtained, considering variations in the number of agents, space complexity, agents' screening capabilities and reinforcement learning.

Original languageEnglish
Title of host publicationWSC 2018 - 2018 Winter Simulation Conference
Subtitle of host publicationSimulation for a Noble Cause
PublisherIEEE
Pages965-976
Number of pages12
ISBN (Electronic)978-1-5386-6572-5
DOIs
Publication statusPublished - 4 Dec 2019
Event2018 Winter Simulation Conference, WSC 2018 - Gothenburg, Sweden
Duration: 9 Dec 201812 Dec 2018
http://meetings2.informs.org/wordpress/wsc2018/

Conference

Conference2018 Winter Simulation Conference, WSC 2018
Abbreviated titleWSC 2018
CountrySweden
CityGothenburg
Period9/12/1812/12/18
Internet address

Fingerprint

Agent-based Model
Reinforcement Learning
Reinforcement learning
Space Complexity
Computational Model
Screening
Learning Algorithm
Update
Learning algorithms
Configuration
Learning
Communication
Model

Cite this

Songhori, M. J., & García-Díaz, C. (2019). Collective problem-solving in evolving networks: An agent-based model. In WSC 2018 - 2018 Winter Simulation Conference: Simulation for a Noble Cause (pp. 965-976). [8632328] IEEE. https://doi.org/10.1109/WSC.2018.8632328
Songhori, Mohsen Jafari ; García-Díaz, César. / Collective problem-solving in evolving networks : An agent-based model. WSC 2018 - 2018 Winter Simulation Conference: Simulation for a Noble Cause. IEEE, 2019. pp. 965-976
@inproceedings{af40d73eb0c04dfda11c6cabb4585678,
title = "Collective problem-solving in evolving networks: An agent-based model",
abstract = "Research works in collective problem-solving usually assume fixed communication structures and explore effects thereof. In contrast, in real settings, individuals may modify their set of connections in the search of information and feasible solutions. This paper illustrates how groups collectively search for solutions in a space under the presence of dynamic structures and individual-level learning. For that, we built an agent-based computational model. In our model, individuals (i) simultaneously conduct search of solutions over a complex space (i.e. a NK landscape), (ii) are initially connected to each other according to a given network configuration, (iii) are endowed with learning capabilities (through a reinforcement learning algorithm), and (iv) update (i.e. create or severe) their links to other agents according to such learning features. Results reveal conditions under which performance differences are obtained, considering variations in the number of agents, space complexity, agents' screening capabilities and reinforcement learning.",
author = "Songhori, {Mohsen Jafari} and C{\'e}sar Garc{\'i}a-D{\'i}az",
year = "2019",
month = "12",
day = "4",
doi = "10.1109/WSC.2018.8632328",
language = "English",
pages = "965--976",
booktitle = "WSC 2018 - 2018 Winter Simulation Conference",
publisher = "IEEE",
address = "United States",

}

Songhori, MJ & García-Díaz, C 2019, Collective problem-solving in evolving networks: An agent-based model. in WSC 2018 - 2018 Winter Simulation Conference: Simulation for a Noble Cause., 8632328, IEEE, pp. 965-976, 2018 Winter Simulation Conference, WSC 2018, Gothenburg, Sweden, 9/12/18. https://doi.org/10.1109/WSC.2018.8632328

Collective problem-solving in evolving networks : An agent-based model. / Songhori, Mohsen Jafari; García-Díaz, César.

WSC 2018 - 2018 Winter Simulation Conference: Simulation for a Noble Cause. IEEE, 2019. p. 965-976 8632328.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Collective problem-solving in evolving networks

T2 - An agent-based model

AU - Songhori, Mohsen Jafari

AU - García-Díaz, César

PY - 2019/12/4

Y1 - 2019/12/4

N2 - Research works in collective problem-solving usually assume fixed communication structures and explore effects thereof. In contrast, in real settings, individuals may modify their set of connections in the search of information and feasible solutions. This paper illustrates how groups collectively search for solutions in a space under the presence of dynamic structures and individual-level learning. For that, we built an agent-based computational model. In our model, individuals (i) simultaneously conduct search of solutions over a complex space (i.e. a NK landscape), (ii) are initially connected to each other according to a given network configuration, (iii) are endowed with learning capabilities (through a reinforcement learning algorithm), and (iv) update (i.e. create or severe) their links to other agents according to such learning features. Results reveal conditions under which performance differences are obtained, considering variations in the number of agents, space complexity, agents' screening capabilities and reinforcement learning.

AB - Research works in collective problem-solving usually assume fixed communication structures and explore effects thereof. In contrast, in real settings, individuals may modify their set of connections in the search of information and feasible solutions. This paper illustrates how groups collectively search for solutions in a space under the presence of dynamic structures and individual-level learning. For that, we built an agent-based computational model. In our model, individuals (i) simultaneously conduct search of solutions over a complex space (i.e. a NK landscape), (ii) are initially connected to each other according to a given network configuration, (iii) are endowed with learning capabilities (through a reinforcement learning algorithm), and (iv) update (i.e. create or severe) their links to other agents according to such learning features. Results reveal conditions under which performance differences are obtained, considering variations in the number of agents, space complexity, agents' screening capabilities and reinforcement learning.

UR - http://www.scopus.com/inward/record.url?scp=85062620969&partnerID=8YFLogxK

U2 - 10.1109/WSC.2018.8632328

DO - 10.1109/WSC.2018.8632328

M3 - Conference contribution

SP - 965

EP - 976

BT - WSC 2018 - 2018 Winter Simulation Conference

PB - IEEE

ER -

Songhori MJ, García-Díaz C. Collective problem-solving in evolving networks: An agent-based model. In WSC 2018 - 2018 Winter Simulation Conference: Simulation for a Noble Cause. IEEE. 2019. p. 965-976. 8632328 https://doi.org/10.1109/WSC.2018.8632328