Do generic interactions copy a system's state redundantly into its environment?
44No. 1,011 in all physics · No. 47 in Quantum Information and Foundations
For a system $S$ coupled to an environment of $N$ fragments by a Hamiltonian $H$, quantum Darwinism requires the mutual information $I(S:F)$ to reach the plateau $H(S)$ (entropy of the pointer observable) for fragments $F$ holding a fraction $f \ll 1$ of the environment, so that the redundancy $R = 1/f$ grows with $N$; the stricter form is a spectrum broadcast structure (fragments hold perfectly distinguishable records correlated with the same outcome). Brandao, Piani and Horodecki (Nature Communications 2015) proved that for any dynamics most fragments can access only one fixed observable of $S$, which does not by itself imply agreement on outcomes. Classify the Hamiltonians (local, with interacting environment fragments, chaotic, at finite temperature $T$) for which $R$ grows linearly with $N$ at late times.
What would settle it
Theorems giving necessary and sufficient conditions on H for redundancy growing with N, checked on exactly solvable spin-bath and photon-scattering models.
Status in the literature
Unverified note: it names recent work and was partly written without a live literature search; it awaits a check by an editor.
Redundancy plateaus were observed in engineered superconducting-qubit settings in 2025 (Zhu et al., Science Advances 11, doi:10.1126/sciadv.adx6857); genericity for natural Hamiltonians is unproved.
No solutions, ideas or subproblems yet.
See also
Details, signature and history
- Kind
- Well-posed question
- Answer
- Classification
- Topic review
- Zurek, Decoherence, einselection, and the quantum origins of the classical, Reviews of Modern Physics, 2003. Link
- Record
- Revision 1 · imported from the seed collection · compiled by AI, not yet reviewed by an editor
Proof
- Fingerprint
- fa5a2dd10678…
- Public log
- Entry 1,989, in checkpoint 2,123
- In the log by
- Block 969,698, 2026-10-03
This text was in the PubPhys log by that date. The date says nothing about when the problem was first posed.
Comments (0)
Comments are not records: they are not signed, stamped or logged. How comments work
No comments yet.
Sign in to comment.