3 ms·
Both the article and the paper mention interventions and refer to Pearl’s work, so I think we can assume this is just a matter of poor phrasing and a child’s ob
by skybrian 6y ago
Both the article and the paper mention interventions and refer to Pearl’s work, so I think we can assume this is just a matter of poor phrasing and a child’s observations of its own interventions aren’t meant to be excluded.
But for the purposes of machine learning, it seems like it should be possible to learn from observations of someone else’s interventions?