> I've seen it happen that a move wasn't considered as a top move by the engine, but once played the engine realizes it's actually crushing. Something about the heuristics used to prune the vast search space can make it miss sacrifices or seemingly sub-optimal moves that temporarily weakens the perceived position but has a huge payoff in the end. But humans find them.
Just to observe, humans display exactly the same phenomenon of ignoring a move before it's made while still being able to realize, after it's made, that it was very strong and ignoring it was a mistake.
OP's choice of words is inexact. The engine has a different intuition, not none. Humans will usually explore trades and overlook seemingly "slow" moves that the computer usually finds.
What the engine does lacks is what chess players call "theory," hence the need for endgame DBs. Humans can also beat engines with "logic," like the Naka match. That one was extraordinary. A simpler & more typical example is isolating a pawn so that it can't ever be defended, but also can't be attacked immediately. For a human, it's easy to understand this is long term vulnerability even if the pawn survives for 20+ more moves.
Anti-computer strategies, ironically, force you to "play the player, not the board."
Exactly. In fact you could argue that intuition, ‘arriving at a conclusion without a conscious understanding of your reasoning’, is the only thing a neural net displays.
Intuitions often do not deliver the correct response; many puzzles and paradoxes (not to mention a few philosophical arguments) hang on misleading intuitions.
Furthermore, heuristics are rules, while intuitions are not. Sometimes people use their intuition to go against what a heuristic suggests!
Just to observe, humans display exactly the same phenomenon of ignoring a move before it's made while still being able to realize, after it's made, that it was very strong and ignoring it was a mistake.