Aydin Mohseni's repositories
Frequency-Dependent-Moran-Process
Computational model of the Moran process for 2x2 symmetric normal form games. Derives the Nash equilibria, evolutionarily stable states, and stochastically stable state of each game, given the size of the population, mutation rate, and intensity of selection. Also, plots the stationary distribution, and allows for the simulation of individual evolving populations.
Bounded-Confidence-Model
This model explores the question how opinion formation within an interacting group leads to consensus, polarization, or fragmentation.
Truth-and-Conformity-on-Networks
This simulation models how social networks can influence the flow and reliability of information in inquiry. In the model, heterogeneous agents--driven, to varying degrees, by the competing priorities of accuracy and of conformity to one’s peers--share public opinions. Agents learn, via Bayesian conditionalization, both from private signal from nature, and from the public opinions of other agents.
EM-Algorithm-for-Gaussian-Mixtures
The estimation maximization algorithm demonstrates an unsupervised, iterative approach to finding maximum likelihood estimates of parameters for a gaussian mixture model, where the model depends on unobserved latent variables.
Minority-Majority-Experiment
A decision-making experiment, created in oTree for the MTurk platform. Eight players are divided into (minority and majority) groups, and are randomly matched for 100 rounds to play a simplified Nash bargaining game.
Reporting-and-Bias-Alpha
This model explore the way that individual confirmation bias along with hyperbole and cherry-picking of events in news media can lead to belief polarization.
Self-Assembling-Networks
This model explores how a social network might self-assemble from the ritualization of the decisions of individual agents. These models show how inquiry in an evolved social network may be significantly more reliable than the individual inquiry of the agents. They also show how such networks may dramatically lower the epistemic risk faced by each inquirer. We consider both models of both perfect and imperfect communication and compare the evolved behavior of inquirers in each.
Causal-Modeling-Playground
A playground for experimenting with some causal Bayes nets.
Dynamic-Network-Widget
Dynamic Network Widget
Multi-Armed-Bandit
Hierarchical and sequential multi-armed bandit model of scientific inquiry, meant to illustrate the problem of unconceived alternatives.
Near-Zero-Sum-Game
A model of Bayesian learning in near-zero-sum games of imperfect information.
Reporting-and-Bias-0
A model of how news reporting interacts with confirmation bias to produce belief distortion.
Reporting-and-Bias-Beta
A model of distortions created by news reporting.
Reporting-and-Bias-Gamma
A model of distortions created by news reporting.
Reporting-Protocols-and-the-Reliability-of-Science
A model of the reliability of scientific findings as affected by the convention threshold for statistical significance and by various methods of reporting findings.
Tragedies-of-the-Commons
A model of AI policy public goods problems under the fermi dynamics.