Reminds me of Giddens and debates around "path dependency" in historical sociology. Bart Kosko has some interesting work on this, if you're willing to go down the fuzzy-logic wormhole. Kosko says that we cannot capture initial states with infinite precision, reflecting the human condition. Digitalization of course exacerbates the problem. This jives with Deweyian pragmatism about the goals and aims of science. As you mention, there is definitely a connection between romantic philosophers and algorithmic information theory. I'm confused by the word "determinism." How does probability relate to determinism? In the debates by philosophers on free will they appear unaware of probability, but I don't think they're that naive.