Probabilities from entanglement, Born's rule pk = ∣ ψk ∣2 from envariance
Abstract
I show how probabilities arise in quantum physics by exploring the implications of environment-assisted invariance or envariance, a recently discovered symmetry exhibited by entangled quantum systems. Envariance of perfectly entangled “Bell-like” states can be used to rigorously justify complete ignorance of the observer about the outcome of any measurement on either of the members of the entangled pair. For more general states, envariance leads to Born’s rule pk∝∣ψk∣2 for the outcomes associated with Schmidt states. The probabilities derived in this manner are an objective reflection of the underlying state of the system—they represent experimentally verifiable symmetries, and not just a subjective “state of knowledge” of the observer. This envariance-based approach is compared with and found to be superior to prequantum definitions of probability including the standard definition based on the “principle of indifference” due to Laplace and the relative frequency approach advocated by von Mises. Implications of envariance for the interpretation of quantum theory go beyond the derivation of Born’s rule: Envariance is enough to establish the dynamical independence of preferred branches of the evolving state vector of the composite system and, thus, to arrive at the environment-induced superselection (einselection) of pointer states, which was usually derived by an appeal to decoherence. The envariant origin of Born’s rule for probabilities sheds light on the relation between ignorance (and, hence, information) and the nature of quantum states.
- Publication:
-
Physical Review A
- Pub Date:
- May 2005
- DOI:
- arXiv:
- arXiv:quant-ph/0405161
- Bibcode:
- 2005PhRvA..71e2105Z
- Keywords:
-
- 03.65.Ta;
- 03.67.-a;
- 02.50.Cw;
- Foundations of quantum mechanics;
- measurement theory;
- Quantum information;
- Probability theory;
- Quantum Physics
- E-Print:
- Figure and an appendix (Born's rule for continuous spectra) added. Presentation improved. (Comments still welcome...)