In complex contexts, simulate

Machine Learning implicitly assumes that a closed form solution exists and uses optimization to approximate an assumed form of it (linear, quadratic etc).

In complex (Complexity) contexts, this might not be true. With more compute, we can do better – Simulate.

Complex systems are computationally irreducible

Notes mentioning this note


Here are all the notes in this garden, along with their links, visualized as a graph. If you don't see any nodes try zooming and panning in the grey area.