Workshop on Agent-Based Modeling (ABM)

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23330%2F19%3A43955365" target="_blank" >RIV/49777513:23330/19:43955365 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Workshop on Agent-Based Modeling (ABM)

Popis výsledku v původním jazyce

Computational methods are the new third pillar of modern science – so besides theoretical work and empirical research we recognize also computer simulations as third way of scientific development. Basically, computer simulations can inspire theoretical work, test feasibility of theoretical ideas, inform empirical research and extend empirical research results. But in all cases, computer simulations answer clearly stated research questions. This introductory workshop – even for students with no experience in programming – focuses on how to use ABM software NetLogo for meaningful carrying-out of computational social science. NetLogo is free software developed at Northwestern University, whole community and all experts agree on NetLogo being the easiest learn computer language for social computer simulations. We will cover whole process, from formulation of research question, through model formulation, model programming, code testing, model validation, and analysis of results. So, the workshop is not about mere programming, but also about meaningful model use. The workshop starts with joint reproducing of 3 classical models, follows with joint realization of one more model and finishes with one week of students’ group work. We will work every working day in two sessions – morning 9AM–12AM and afternoon 2PM–5PM – since Monday 6/17 until Friday 6/28 (schedule is described in detail below). We will cover whole process since research question formulation through formulating conceptual model, writing ODD documentation, programing of model, calibration and validation, finishing with experimentation, analysis and presentation of results. This process we will go through several times – jointly with each classical model and then separately in working groups, each working group with its own project. After two weeks of workshop the student is expected to have experience with reproducing 4 models in joint work and deep experience with creating, experimenting and analyzing her own model in working group. Regarding preparation for the workshop, students are expected to bring their own laptops with software NetLogo and R installed, they should also read papers accompanying classical models, and think about process suitable for modelling. R and NetLogo are free software, students are expected to download and install them on their own laptops on their own. Also RStudion, which is beneficial for working with R is free for download and use. Pages for download are here: https://ccl.northwestern.edu/netlogo/download.shtml and https://cran.r-project.org/ and https://www.rstudio.com/products/rstudio/download/. Students are also expected to open at least three models from NetLogo library (in NetLogo software) and play with these models. They should also try to play with R, there is training script in the cloud storage: https://owncloud.cesnet.cz/index.php/s/werCFgzXSxgfLBW. This storage contains also papers describing three classical models (Schelling 1971 – segregation; Nowak et al. 1990 – dynamic social impact; Axelrod 1997 – culture) and one more joint model (Janssen, Jäger 2002; Bravo et al. 2013). These papers are not needed to be studied in detail, but just briefly to get comprehension what is the modeled process and how it is modeled. There is also paper Grimm et al. 2010 which describes ODD documentation and the first part of Railsback and Grimm 2011 book focused on developing a simple model including ODD protocol. Students are expected to read this paper and the part of book carefully – after the workshop will be clear that creation of good ODD documentation is crucial step in modeling process and. In the storage there are also some further reading materials, but their study is not mandatory. The last thing for the preparation is to think about own problem for modeling. After reading of materials it might be easier to think it out, students also have to formulate their definite research question at Friday 6/21, so they have time during the first week of workshop, but we will start discussing the research questions and topics for modelling since the very first day of workshop. Ideal processes for the modelling usually fulfill these criteria: (a) agents are autonomous, not ruled by central authority, (b) agents decide how to act according their environment (typically other agents), and (c) agents’ decisions change environment (e.g. I decide to sell stock if majority of other agents wants to buy it, so my decision buy or sell changes size of selling and buying group).

Název v anglickém jazyce

Workshop on Agent-Based Modeling (ABM)

Popis výsledku anglicky

Computational methods are the new third pillar of modern science – so besides theoretical work and empirical research we recognize also computer simulations as third way of scientific development. Basically, computer simulations can inspire theoretical work, test feasibility of theoretical ideas, inform empirical research and extend empirical research results. But in all cases, computer simulations answer clearly stated research questions. This introductory workshop – even for students with no experience in programming – focuses on how to use ABM software NetLogo for meaningful carrying-out of computational social science. NetLogo is free software developed at Northwestern University, whole community and all experts agree on NetLogo being the easiest learn computer language for social computer simulations. We will cover whole process, from formulation of research question, through model formulation, model programming, code testing, model validation, and analysis of results. So, the workshop is not about mere programming, but also about meaningful model use. The workshop starts with joint reproducing of 3 classical models, follows with joint realization of one more model and finishes with one week of students’ group work. We will work every working day in two sessions – morning 9AM–12AM and afternoon 2PM–5PM – since Monday 6/17 until Friday 6/28 (schedule is described in detail below). We will cover whole process since research question formulation through formulating conceptual model, writing ODD documentation, programing of model, calibration and validation, finishing with experimentation, analysis and presentation of results. This process we will go through several times – jointly with each classical model and then separately in working groups, each working group with its own project. After two weeks of workshop the student is expected to have experience with reproducing 4 models in joint work and deep experience with creating, experimenting and analyzing her own model in working group. Regarding preparation for the workshop, students are expected to bring their own laptops with software NetLogo and R installed, they should also read papers accompanying classical models, and think about process suitable for modelling. R and NetLogo are free software, students are expected to download and install them on their own laptops on their own. Also RStudion, which is beneficial for working with R is free for download and use. Pages for download are here: https://ccl.northwestern.edu/netlogo/download.shtml and https://cran.r-project.org/ and https://www.rstudio.com/products/rstudio/download/. Students are also expected to open at least three models from NetLogo library (in NetLogo software) and play with these models. They should also try to play with R, there is training script in the cloud storage: https://owncloud.cesnet.cz/index.php/s/werCFgzXSxgfLBW. This storage contains also papers describing three classical models (Schelling 1971 – segregation; Nowak et al. 1990 – dynamic social impact; Axelrod 1997 – culture) and one more joint model (Janssen, Jäger 2002; Bravo et al. 2013). These papers are not needed to be studied in detail, but just briefly to get comprehension what is the modeled process and how it is modeled. There is also paper Grimm et al. 2010 which describes ODD documentation and the first part of Railsback and Grimm 2011 book focused on developing a simple model including ODD protocol. Students are expected to read this paper and the part of book carefully – after the workshop will be clear that creation of good ODD documentation is crucial step in modeling process and. In the storage there are also some further reading materials, but their study is not mandatory. The last thing for the preparation is to think about own problem for modeling. After reading of materials it might be easier to think it out, students also have to formulate their definite research question at Friday 6/21, so they have time during the first week of workshop, but we will start discussing the research questions and topics for modelling since the very first day of workshop. Ideal processes for the modelling usually fulfill these criteria: (a) agents are autonomous, not ruled by central authority, (b) agents decide how to act according their environment (typically other agents), and (c) agents’ decisions change environment (e.g. I decide to sell stock if majority of other agents wants to buy it, so my decision buy or sell changes size of selling and buying group).

Druh

W - Uspořádání workshopu

CEP obor

—

OECD FORD obor

50401 - Sociology

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2019

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Místo konání akce

Marakéš, Maroko

Stát konání akce

MA - Marocké království

Datum zahájení akce

—

Datum ukončení akce

—

Celkový počet účastníků

6

Počet zahraničních účastníků

—

Typ akce podle státní přísl. účastníků

CST - Celostátní akce