3 ms·
His argument is more nuanced, and ... includes examples. I've only managed about 3 chapters of the book so far. It has proved quite illuminating. There are e
by dredmorbius 5y ago
His argument is more nuanced, and ... includes examples.
I've only managed about 3 chapters of the book so far. It has proved quite illuminating.
There are examples of systems that do model well and those that don't. Depending on your physics, the 3, 2, 1, or null-body problems. The double pendulum. Etc.
And those are only at the very simplest level.
The fact that some systems can be effectively modelled doesn't mean that all can be. And in the case of complex systems (e.g., the German "rational forestry" method described early in Scott's book), initial success may preceed subsequent catastrophe.
- 9wzYQbTYsAIc 5y ago> The fact that some systems can be effectively modelled doesn't mean that all can be. Based on my understanding, the leading-edge of scientific research shows that “if the human mind can comprehend it, then there is a mathematical function for it” Are you positing, then, that there simply are limits to what humans can understand? Or, rather, that some systems are just too complex to document properly?
- dredmorbius 5y ago1. No. 2. Yes, yes.
- 9wzYQbTYsAIc 5y ago1. http://neuralnetworksanddeeplearning.com/chap4.html http://neuralnetworksanddeeplearning.com/chap4.html, https://mitpress.mit.edu/books/parallel-distributed-processing-volume-1 https://mitpress.mit.edu/books/parallel-distributed-processi... 2. https://en.wikipedia.org/wiki/Scientific_skepticism https://en.wikipedia.org/wiki/Scientific_skepticism
- dredmorbius 5y ago1. Halting problem, as a reductionist example. 2. I don't follow your point. I do appreciate the examples/links.