An active area of research in the fields of machine learning and statisticsis the development of causal discovery algorithms, the purpose of which is toinfer the causal relations that hold among a set of variables from thecorrelations that these exhibit. We apply some of these algorithms to thecorrelations that arise for entangled quantum systems. We show that they cannotdistinguish correlations that satisfy Bell inequalities from correlations thatviolate Bell inequalities, and consequently that they cannot do justice to thechallenges of explaining certain quantum correlations causally. Nonetheless, byadapting the conceptual tools of causal inference, we can show that any attemptto provide a causal explanation of nonsignalling correlations that violate aBell inequality must contradict a core principle of these algorithms, namely,that an observed statistical independence between variables should not beexplained by fine-tuning of the causal parameters. In particular, wedemonstrate the need for such fine-tuning for most of the causal mechanismsthat have been proposed to underlie Bell correlations, including superluminalcausal influences, superdeterminism (that is, a denial of freedom of choice ofsettings), and retrocausal influences which do not introduce causal cycles.
展开▼
