I like the idea of separating optimization from agency! But I am confused by your description of optimization. What you seem to say is it is “non-random”. Does non-random mean “losslessly compressible”? Or lossily compressible? Or compressible with some specific compressor? Is it relative to some entity doing the compressing?
These are good questions that I probably have stuff to say about, although I’m not entirely clear enough on your question to know exactly what responses to give.
I spend most of my time thinking in terms of deterministic systems, so I was thinking losslessly compressible via K-complexity (which is a specific compressor (modulo choice of universal Turing machine) which is not relative to the entity doing the compressing (though I’m chewing over further thoughts related to that)).
I like the idea of separating optimization from agency! But I am confused by your description of optimization. What you seem to say is it is “non-random”. Does non-random mean “losslessly compressible”? Or lossily compressible? Or compressible with some specific compressor? Is it relative to some entity doing the compressing?
These are good questions that I probably have stuff to say about, although I’m not entirely clear enough on your question to know exactly what responses to give.
I spend most of my time thinking in terms of deterministic systems, so I was thinking losslessly compressible via K-complexity (which is a specific compressor (modulo choice of universal Turing machine) which is not relative to the entity doing the compressing (though I’m chewing over further thoughts related to that)).