Gary Marcus

i am not endorsing the specifics of his technical machinery; i think getting AI to reason about causality is at best a partly solved problem, with many remaining mysteries. i am reacting to your assimilation of causation and correlation.

So what definition of causality do you currently use? Pearl's definition relies on his machinery, interventions. If you're implying that we can't apply a notion of causality to AI in a straightforward way, perhaps you should recognise that a possible answer may be that.. 1/

a bag of tricks/heuristics (running away from predators even when one has never been attacked, or the presence of newborn reflexes) may be sufficient for an agent/AI. When I said "We probably just correlate things to the point we believe A causes B.", .. 2/

all I meant is that while we have a different definition of correlation and causation in different frameworks (e.g., Pearl's), this *does not* imply that we need it to build AIs. 3/3

just don’t see what you are arguing for.

Let me try again, what do you define as causality? If you think Pearl's definition is useful, how do you use it to build AIs? If you use another definition, which one? If you are not sure about the definition itself, why do you think that causality is what we need to build AIs?

i wrote about definitions yesterday. i can't define a chair, but i can sit in one; i can't define a game, but I can play one. Generally haven't found definitional discussions to be useful, outside mathematics. .

Ok, you say we need "causality" (among other things). Can you operationalise "causality" so that I can start building an AI according to your proposal?