arXiv:2610.25196
Exact probability laws of random inner products: Dirichlet splines, overlaps, uncertainty, and Bargmann invariants
Lin Zhang
Abstract
Inner products of random matrices and random quantum states---such as orbital inner products, expectation values, overlaps, and cyclic Bargmann invariants---are shown to follow from two underlying mechanisms: the diagonal of an induced random state is Dirichlet distributed, and the expectation value of a random pure state in a rank-one or two-by-two operator is a numerical shadow on an elliptical disk. For states induced by an environment of dimension $K$, the expectation value of any observable is a Curry--Schoenberg spline with each eigenvalue repeated $K$ times, and the vector of moments has an explicit density on a simplex; for a qutrit, the variance law is obtained in finite form. The overlap of two random qubit states attains its mode at its mean for every rotation-invariant ensemble, but for Hilbert--Schmidt qutrits the density at the mean has a strictly negative derivative, so mode and mean separate; for large environments, the centred overlap converges to a variance-gamma law rather than a Gaussian. For two noncommuting qubit observables, the joint law of the two standard deviations is derived for all $K$, and the probability of coming within a small distance of the optimal sum of variances scales as a power of that distance, with a different exponent when the Bloch axes are orthogonal. For three Haar-random qubit states, the exact density of the Bargmann invariant is given in terms of a complete elliptic integral, together with its local behaviour and elementary marginals. In every dimension, the probability that its real part is negative---certifying basis-independent coherence---is computed, along with all mixed moments. For cycles of arbitrary length, the invariant is shown to be driven by an exact three-parameter Markov chain, and the logarithm of its modulus obeys a law of large numbers and a central limit theorem.
AI Summary
The paper studies exact probability laws, rather than attainable ranges, for orbital inner products, expectation values, overlaps, uncertainty quantities, and cyclic Bargmann invariants of random quantum states. Its main tools are the Dirichlet distribution of the diagonal of an induced state $\rho\sim\mu_{d,K}$ and numerical-shadow distributions on ellipses for rank-one or two-dimensional operators. Expectation values $\operatorname{Tr}(A\rho)$ are expressed as B-splines with each eigenvalue repeated $K$ times; moment vectors have explicit Dirichlet-derived densities on simplices. A finite-form variance density is obtained for the qutrit observable $A=\operatorname{diag}(-1,0,1)$. Independent rotation-invariant qubit ensembles have overlap mode at the mean $1/2$, whereas Hilbert–Schmidt qutrit overlaps have density derivative approximately $-4.815$ at their mean $1/3$. At fixed dimension and large environment sizes, scaled overlap fluctuations converge to a variance-gamma, not Gaussian, distribution via Wishart fluctuations. For two qubit observables, explicit joint uncertainty densities yield near-optimal sum-of-variance probabilities scaling as $\varepsilon^{K-1/2}$ for orthogonal Bloch axes and $\varepsilon^K$ otherwise. For three Haar-random qubit states, the Bargmann-invariant density involves a complete elliptic integral. In dimension $d$, its negative-real-part probability is $1/2-\sqrt\pi\Gamma(d)/[4\Gamma(d+1/2)]$. An exact three-real-parameter Markov chain handles arbitrary cycle lengths and yields logarithmic-modulus laws of large numbers and central limit theorems. Stated open questions include uniqueness of the qutrit overlap mode, higher-dimensional Bargmann-density singularities and boundary behavior, and uncertainty laws for noncommuting qutrit observables.
Generative-AI tools were used for language refinement and assistance with computational checks.
Error found: The abstract's spline-density claim needs to exclude scalar observables A=aI, whose expectation-value law is the point mass δ_a. The proof of Theorem b3density misstates the major semi-axis and temporarily drops a factor 1/√u from its integral, although the subsequent quartic integral and final density are correct.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 64 pages,7 figures