Aha, thanks for that extremely useful clarification! I was also confused from the abstract and intro about exactly what the inputs and outputs were. Yes, the clear impression was created that they learned the rules---in the sense that they could then generate moves that would (usually? almost always?) obey the rules---when it sounds like really they just learned an evaluation function that encodes lots of implicit knowledge about the rules, of course, but that still requires a piece of code that knows the rules explicitly as a "guardrail" when generating moves.
Since you obviously know this subject, what are the best current results in the direction of inducing the rules of chess from played games?
Since you obviously know this subject, what are the best current results in the direction of inducing the rules of chess from played games?