Connections between AI and ecology & biosemiotics

Connections between AI and ecology & biosemiotics
======

Historically, findings in AI has contributed to other fields. Formulas in reinforcement learning was picked up by neuroscientists to explain how dopamine is released and how it affects learning.These examples provide a way - a hope for me to find satisfaction for that specific condition also add which so often I see as engineering rather than science.

Since I'm curious about ecology, particularly the relationship between environments and agents; and biosemiotics, how communication and language comes about, I went ahead and looked at these with AI assistance.

Fields in AI I should pay attention to, in relation to ecology:
- Model-based RL: Model-free has a fixed world model. Model-based learns the world model. What if it can also learn how to state evolves? For example, climate change.
	- Look into non-stationary environments, and non-episodic environments
	- What if we model the environment as an agent?
- Multi-Agent Stigmergy
- "Niche Construction" and its relationship with agent optimization. See https://pmc.ncbi.nlm.nih.gov/articles/PMC11606325/
- Homeostatic Reinforcement Learning
- Malthusian Reinforcement Learning (MRL)
- Ecological Reinforcement Learning (EcoRL)

For biosemiotics I'm not sure yet.
Maybe linguistics link to it?