Archive

Formalising what kinds of active inference agents closure of constraints picks out
Investigating properties of this class of agents
Finding a tractable principle equivalent to closure of constraints, that can be used to detect this property in systems where the processes are too numerous to model explicitly
What does it mean for something to be alive? How is this distinct from something being agentic - how does aliveness affect agency? When should we expect things to be alive? Related
Must enactivism and FEP fight to the death or are they just dual? Wouldn't it be nice if they were friends...

Finishing The Second FEP Explainer Before It Finishes Me
Backburner: control, blanket expansion, love
A World Beyond Physics: The Emergence and Evolution of Life by Stuart Kauffman
Chess, irl
did I mention exams

Relative to an observer, some picks of Markov blanket are better than others. More precisely, some picks of Markov blanket trajectories are better than others. Why do these better picks of blanket (trajectories) change?
Control has something to do with it. But what does it mean for one thing to control another thing? Self-representation has something to do with it. But what does it mean for a thing to represent itself as a thing, or represent other things as things? Love has something to do with it. Oh great, how simple. Love. That's it. Just love.
Given that the world is on fire, what do I want to do about it? How do I want to live?
Why is it so hard to find a decent ramen shop in London?
Analysis I by Terence Tao
Emerging out of my research hole, slowly. Meatbag to meatbag, talking to humans, slowly.