A quantum Bayes’ rule and related inference

Mohit Lal Bera, Manabendra Nath Bera

The traditional Bayes' rule lays the foundation for causal reasoning and finding relations between cause (input) and effect (output). This causal reasoning is universally applied to all physical processes to establish causal relations. Here we show that it does not establish correct causal correspondence between quantum causes and effects in general. In fact, there are instances within the framework of quantum mechanics where the use of the traditional Bayes' rule leads to inconsistencies in quantum measurement inferences. We consider two such cases, inspired by Frauchiger-Renner's and Hardy's setups, where the traditional Bayes' rule results in paradoxical situations even after assuming quantum mechanics as a nonlocal theory. As a remedy, we introduce an input-output causal relation using the reasoning based on a quantum Bayes' rule. It applies to general quantum processes even when a cause (or effect) is in coherent superposition with other causes (or effects), involves nonlocal correlations as allowed by quantum mechanics, and applies in the cases where causes belonging to one system induce effects in some other system as happens in quantum measurement processes. This enables us to propose a resolution to the contradictions that appear in the context of Frauchiger-Renner's and Hardy's setups. Our results thereby affirm that quantum mechanics, equipped with a quantum Bayes' rule, can indeed consistently describe the use of itself.

Hans Christian von Baeyer

Jon Frost, Carlos Madeira, Harald Uhlig

Mohit Lal Bera, Manabendra Nath Bera

Classical Bayes' rule lays the foundation for the classical causal relation between cause (input) and effect (output). This causal relation is believed to be universally true for all physical processes. Here we show, on the contrary, that it is inadequate to establish correct correspondence between cause and effect in quantum mechanics. In fact, there are instances within the framework of quantum mechanics where the use of classical Bayes' rule leads to inconsistencies in quantum measurement inferences, such as Frauchiger-Renner's paradox. Similar inconsistency also appears in the context of Hardy's setup even after assuming quantum mechanics as a non-local theory. As a remedy, we introduce an input-output causal relation based on quantum Bayes' rule. It applies to general quantum processes even when a cause (or effect) is in coherent superposition with other causes (or effects), involves nonlocal correlations as allowed by quantum mechanics, and in the cases where causes belonging to one system induce effects in some other system as it happens in quantum measurement processes. This enables us to propose a resolution to the contradictions that appear in the context of Frauchiger-Renner's and Hardy's setups. Our results thereby affirm that quantum mechanics, equipped with quantum Bayes' rule, can indeed consistently explain the use of itself.

Rüdiger Schack, Todd A. Brun, Carlton M. Caves

We state a quantum version of the Bayes rule for statistical inference and give a simple general derivation within the framework of generalized measurements. The rule can be applied to measurements on N copies of a system if the initial state of the N copies is exchangeable. As an illustration, we apply the rule to N qubits. Finally, we show that quantum state estimates derived via the principle of maximum entropy are fundamentally different from those obtained via the quantum Bayes rule.

Christopher G. Timpson