Proposal: Integrating an AI-Mediated Coalition Formation Algorithm

[HILLELOH]

Hi there,

I am currently working on an academic project focusing on already published article: Computational Social Choice, and I came across your excellent comchoice library. I would love to contribute by implementing a recent algorithm and wanted to ask for your thoughts on whether it fits your repository's scope.

About the Algorithm:
The algorithm is based on the recent paper "AI-Generated Compromises for Coalition Formation" (Briman, Shapiro, Talmon). It deals with spatial voting and iterative coalition formation.
Given a set of agents with ideal points in a metric space (e.g., 2D Euclidean) and a status quo, an "AI Mediator" iteratively computes the centroid of current coalitions and proposes new compromise points. Agents then vote on whether to join the new coalition based on their distance to the compromise versus the status quo.

The Contribution:
I plan to implement the core logic of this mediator-based coalition formation in Python (starting with the 2D Euclidean space and deterministic agents).

My Questions:

1. Does this iterative coalition formation model fit within the vision of comchoice?
2. If so, do you have any recommendations on where this should live within the repository's architecture?

I am willing to do all the coding, writing the tests, and documentation. I'd be happy to open a PR once I have your green light.

Thanks in advance for your time and for maintaining this project!