# 🧠 Shared Protentions in Multi-Agent Active Inference

**Mahault Albarracin, Riddhi J. Pitliya, Toby St. Clere Smithe, Daniel Ari Friedman, Karl Friston, Maxwell J.D. Ramstead** (2024) · *Entropy*

[![DOI](https://img.shields.io/badge/DOI-10.3390%2Fe26040303-blue)](https://doi.org/10.3390/e26040303)

---
<!-- Schema.org structured data for search engines -->
<!--
{"@context":"https://schema.org","@type":"ScholarlyArticle","headline":"SharedProtentions","abstract":"We develop the concept of shared protentions\u2014shared anticipatory states\u2014in multi-agent Active Inference. Protentions are future-directed expectations that shape perception and action; here we formalize how agents can develop shared protentions through coupled generative models, enabling coordinated anticipatory behavior in multi-agent systems.","keywords":["shared protentions","multi-agent","Active Inference","anticipation","coupled generative models","coordination","predictive processing"],"author":{"@type":"Person","name":"Daniel Ari Friedman","url":"https://danielarifriedman.com/"}}
-->


## Abstract

> This paper integrates concepts from Husserlian phenomenology, active inference, and category theory to create a framework for understanding social action based on shared goals. It emphasizes the role of shared protentions among agents and their connection to collective goal-directed behavior.

## Keywords

`shared protentions` · `multi-agent` · `Active Inference` · `anticipation` · `coupled generative models` · `coordination` · `predictive processing`

## Key Contributions

- Development of a comprehensive framework linking phenomenology, active inference, and category theory.
- Formalization of shared goals in terms of shared protentions within multi-agent systems.
- Introduction of mathematical tools from category theory to model interactions among agents.
- Exploration of the emergence of group intentionality through shared protentions.

## Methods

- Review of Husserlian phenomenology focusing on inner time-consciousness concepts.
- Application of active inference as a formal approach to model agent behavior.
- Utilization of sheaf and topos theory to represent agent interactions and shared understanding.
- Construction of a 'consensus topos' to analyze social action based on shared goals.

## 🎯 Consulting & Tutoring

**Available for AI Research Consulting and Tutoring.** [Contact Daniel Ari Friedman, PhD](https://danielarifriedman.com/) for collaboration on Active Inference, Bayesian modeling, and computational biology.

## Citation

```bibtex
@article{2024_SharedProtentions,
  author = {Mahault Albarracin, Riddhi J. Pitliya, Toby St. Clere Smithe, Daniel Ari Friedman, Karl Friston, Maxwell J.D. Ramstead},
  title = {{Shared Protentions in Multi-Agent Active Inference}},
  journal = {Entropy},
  year = {2024},
  doi = {10.3390/e26040303},
}
```

## File Inventory

- `AGENTS.md` (1,935 bytes)
- `2024_SharedProtentions.pdf` (325,880 bytes)
- `README.md` (1,662 bytes)
- `SKILL.md` (1,894 bytes)