When we isolate fundamental particles, carefully orchestrated experiments may show predictable distributions of outcomes. On atom and molecular level, chemistr

When we isolate fundamental particles, carefully orchestrated experiments may show predictable distributions of outcomes. On atom and molecular level, chemistry also provides distributions of predictability and even more outcome certainty. This follows the general pattern of interesting chaotic systems, and the side-effects of "chaotic attractors". We find areas of stability, where big numbers tend to converge to stable and predictable values/surfaces. We also find chaos and unpredictability between the stable surfaces in outcomes. Without full information, we can approximate using ML, though how they generalize and explain could also be an iterative ML task. The 3-body problem means that even on macro scale we run into problems calculating some orbits (ie. asteroids). It's just that over time, most chaotic orbits tend to stabilize, so we don't run into this too much. With full information, everything can be calculated theoretically. So it is both the fundamental problem of the differential calculus itself, though inaccurate information compounds the fundamental. Virtual particles sound like constructs bridging some of the gaps, though still unsolved on most 3+ body problems. Maybe it is correct to assume they "work" for stable/semi-stable areas of simple models, but break down in between states of chaotic systems? I sense that we search for simple foundational relationships and understandable constants. However, the problem itself seems intractable using reductionism, the more you seek to encompass the whole across scales. Ie. given two random real numbers between 1.0 and 10.0, human beings expect to find integers. We seek to construct perfection, while blinded to the environment sustaining us. The expectation is not totally unwarranted, if you look at the distribution of 2+ random numbers.