Computational Models of Diversity and Inequity
MCMP Summer School, July 22-26, 2024
Notes:
Here are publicly sharable resources from the MCMP Summer School 2024, lecture stream "Computational Models of Diversity and Inequity." This lecture stream consists of four 1.5-hour lectures, and two 50-min exercise sessions. Most of the interactive activities are not shared here due to student privacy. Some of the slides are a bit contextual to the venue I am teaching at, so feel free to skip them.
If you are a student from the summer school, you should have access to another link with more information. Email me if you cannot locate it.
If you are a colleague and would like to use the materials in your teaching, please credit me according to CC BY 4.0.
I received positive feedback on this course and I would love to teach similar materials as a short course in other institutions. Email me if you would like to discuss more.
Course Description:
How do social identities and social injustice impact the production of knowledge in a group? Is social and cognitive diversity beneficial in group learning? If so, in what ways? These are long-standing questions that interest feminist epistemologists, philosophers of science, and social scientists. In this course, we will focus on some recent methodological developments in answering those questions. We will look at a range of philosophical work that use computational agent-based models (ABM) to explore social dynamics, paying special attention to the possible causes and consequences of diversity and inequity.
At the core of those models is a simple idea that social phenomena evolve over time, as a result of agents’ individual behaviours and patterns of interaction with others in the community. If we are interested in what happens if one social group devalues evidence from another social group, for instance, then we can construct simulation models to explore how this discriminatory treatment may evolve to impact the epistemic lives of the community over time.
Over a series of lectures, we will dissect individual components of an ABM, and look at a range of examples to understand how different components of a model come together to solve philosophically and socially relevant questions. We will also explore the trade-offs involved in constructing a model. At the end of the course, students will have the opportunity to sketch the dynamics of an original model in a group.
Topics may include: epistemic injustice, gender and racial credit gaps, gender and racial belief gaps, collaboration inequity in science, mechanisms of diversity of practice in science.
No prior knowledge of coding is required or expected.
Slides:
Lecture 1 (Intro)
Lecture 2 (Primer on ABMs)
Lecture 3 (Bandit Models)
Lecture 4 (From Pseudo-models to Computational Models)
Useful Links:
For model visualization and getting started at coding your first model in Python (very beta): Modelpy
Python learning resources:
Relevant Readings:
Philosophy of Agent-Based Modeling/Overviews:
- Smaldino 2017, "Models are Stupid, and We Need More of Them"
- Šešelja 2023, "Agent-Based Modeling in the Philosophy of Science"
- Mayo-Wilson and Zollman 2021, "The computational philosophy: simulation as a core philosophical method"
- Wu and O'Connor 2021, "How should we promote transient diversity in science?"
Modeling Papers:
Bandit Problem:
- Zollman 2007, "The communication structure of epistemic communities"
- Zollman 2010, "The epistemic benefit of transient diversity"
- Wu 2022 [2023], "Epistemic advantage on the margin: A network standpoint epistemology"
- Fazelpour and Steel 2022, "Diversity, trust, and conformity: A simulation study"
MCMP Summer School, July 22-26, 2024
Notes:
Here are publicly sharable resources from the MCMP Summer School 2024, lecture stream "Computational Models of Diversity and Inequity." This lecture stream consists of four 1.5-hour lectures, and two 50-min exercise sessions. Most of the interactive activities are not shared here due to student privacy. Some of the slides are a bit contextual to the venue I am teaching at, so feel free to skip them.
If you are a student from the summer school, you should have access to another link with more information. Email me if you cannot locate it.
If you are a colleague and would like to use the materials in your teaching, please credit me according to CC BY 4.0.
I received positive feedback on this course and I would love to teach similar materials as a short course in other institutions. Email me if you would like to discuss more.
Course Description:
How do social identities and social injustice impact the production of knowledge in a group? Is social and cognitive diversity beneficial in group learning? If so, in what ways? These are long-standing questions that interest feminist epistemologists, philosophers of science, and social scientists. In this course, we will focus on some recent methodological developments in answering those questions. We will look at a range of philosophical work that use computational agent-based models (ABM) to explore social dynamics, paying special attention to the possible causes and consequences of diversity and inequity.
At the core of those models is a simple idea that social phenomena evolve over time, as a result of agents’ individual behaviours and patterns of interaction with others in the community. If we are interested in what happens if one social group devalues evidence from another social group, for instance, then we can construct simulation models to explore how this discriminatory treatment may evolve to impact the epistemic lives of the community over time.
Over a series of lectures, we will dissect individual components of an ABM, and look at a range of examples to understand how different components of a model come together to solve philosophically and socially relevant questions. We will also explore the trade-offs involved in constructing a model. At the end of the course, students will have the opportunity to sketch the dynamics of an original model in a group.
Topics may include: epistemic injustice, gender and racial credit gaps, gender and racial belief gaps, collaboration inequity in science, mechanisms of diversity of practice in science.
No prior knowledge of coding is required or expected.
Slides:
Lecture 1 (Intro)
Lecture 2 (Primer on ABMs)
Lecture 3 (Bandit Models)
Lecture 4 (From Pseudo-models to Computational Models)
Useful Links:
For model visualization and getting started at coding your first model in Python (very beta): Modelpy
Python learning resources:
- To get started: https://www.youtube.com/watch?v=kqtD5dpn9C8
- More advanced topics: https://www.youtube.com/watch?v=_uQrJ0TkZlc&t=372s
Relevant Readings:
Philosophy of Agent-Based Modeling/Overviews:
- Smaldino 2017, "Models are Stupid, and We Need More of Them"
- Šešelja 2023, "Agent-Based Modeling in the Philosophy of Science"
- Mayo-Wilson and Zollman 2021, "The computational philosophy: simulation as a core philosophical method"
- Wu and O'Connor 2021, "How should we promote transient diversity in science?"
Modeling Papers:
Bandit Problem:
- Zollman 2007, "The communication structure of epistemic communities"
- Zollman 2010, "The epistemic benefit of transient diversity"
- Wu 2022 [2023], "Epistemic advantage on the margin: A network standpoint epistemology"
- Fazelpour and Steel 2022, "Diversity, trust, and conformity: A simulation study"