One thing to note is the authors mostly have "AI for science" in mind, instead of machine learning in general. Quite a few examples in the paper are about disco

One thing to note is the authors mostly have "AI for science" in mind, instead of machine learning in general. Quite a few examples in the paper are about discovering new conservation laws.