Quantum Systems’ Randomness Limited by Initial State, Not Inherent Rules

A thorough investigation into quantum systems with symmetries such as SU reveals key limitations to generating truly random states. Yuhan Wu and Joaquin F. Rodriguez-Nieva, at the Texas A&M University, show that the degree of Haar-like randomization achievable under unitary dynamics is sharply constrained by experimental limitations on state initialisation, particularly low-entanglement initial states. The study demonstrates that time-evolved states can, in principle, reproduce Haar-like behaviour at the level of finite statistical moments, those accessible under realistic experimental conditions with a finite number of state copies. Restrictions on accessible regions of Hilbert space impede the generation of pure random states, addressing obstructions to full randomization.

Initial state constraints limit achievable quantum randomness

Entanglement entropy, a measure of quantum entanglement quantifying the correlation between quantum particles, now reaches a maximum of 0.25L, a substantial decrease from the 0.33L expected for fully random states. This threshold signifies a fundamental limit to achieving Haar-like randomization, the process of generating truly random quantum states, within systems governed by SU symmetry. The primary obstruction is the use of unentangled states during initial state preparation, rather than the symmetry itself. Understanding this distinction is crucial, as symmetries are inherent properties of many physical systems, while initial state preparation is a controllable aspect of experimental design. The SU symmetry group, a mathematical framework describing transformations that leave the system’s Hamiltonian unchanged, dictates the allowed states and transitions, but does not inherently prevent randomization if the initial state allows for sufficient exploration of the Hilbert space. The Hilbert space represents all possible states of the quantum system.

A maximum value of 0.25L was revealed by analysis of entanglement entropy, sharply lower than the 0.33L expected from fully random states, indicating a clear deviation from complete randomization. This difference, while seemingly small, is statistically significant and points to a systematic limitation in the achievable randomness. The calculation of entanglement entropy involves tracing over degrees of freedom, effectively measuring the correlations between subsystems. A value of 0.33L represents the maximum entanglement entropy for a completely random state in this context, serving as a benchmark for comparison. Even with asymptotically long evolution times and strong quantum chaos, late-time states retain distinguishable characteristics, exhibiting finite deviations from complete randomness when assessed via entanglement entropy. Quantum chaos, in this context, refers to the sensitivity of the system’s evolution to initial conditions, analogous to classical chaos but governed by the rules of quantum mechanics. The persistence of these deviations suggests that the initial state’s limitations are not simply ‘washed out’ by time evolution.

Further investigation showed that time-evolved states can, in principle, mimic Haar-like behaviour if the initial state aligns with the statistical moments of conserved operators. Conserved operators are physical quantities that remain constant during the system’s evolution, such as energy or momentum. Their statistical moments provide information about the distribution of these quantities in the system. Alignment with these moments implies that the initial state possesses the necessary structure to evolve towards a random state, given the system’s symmetries. However, commonly used unentangled initial states prevent this alignment. Consequently, prolonged evolution under strong quantum chaos does not eliminate these distinguishable characteristics. The lack of entanglement restricts the system’s ability to explore the full range of states allowed by the conserved operators. While these findings quantify achievable randomness, they do not yet clarify how to overcome initial state limitations to reach truly Haar-random behaviour in practical quantum devices, and future work will focus on strategies for initial state engineering, potentially involving techniques to create more highly entangled initial states.

Initial quantum state preparation governs achievable randomness levels

Generating genuinely random quantum states is increasingly relied upon by scientists for applications ranging from benchmarking new quantum computers to simulating complex materials. Quantum computers require reliable sources of randomness for tasks such as Monte Carlo simulations and key generation in quantum cryptography. Accurate simulations of materials, particularly those with strong correlations, also benefit from the ability to generate random initial conditions. This work clarifies that achieving this randomness isn’t simply about the inherent properties of a quantum system, but is heavily influenced by how those systems are initially prepared. The team’s findings reveal a surprising limitation: commonly used, low-entanglement starting states actively prevent systems from fully exploring the potential for randomness, even when subjected to chaotic dynamics. This highlights the importance of considering the entire experimental setup, not just the quantum system itself.

This work does not invalidate the pursuit of quantum randomness; instead, it refines our understanding of how to achieve it. While initial state preparation sharply impacts the ultimate degree of randomness, the research identifies conditions under which even limited starting states can approach genuinely random behaviour at a measurable level. This is important for practical quantum technologies where perfect initial states are unattainable, allowing for improved experimental design and data interpretation. The limitations on generating truly random quantum states stem from experimental initialisation, not the symmetries inherent within them. This suggests that focusing on improving initial state preparation techniques, rather than attempting to circumvent the system’s symmetries, is a more fruitful avenue for research.

States possessing minimal quantum entanglement, a correlation between particles, prevent full exploration of potential randomness, even when the system behaves chaotically. The team demonstrated that late-time states, even in strongly chaotic systems, exhibit finite and measurable differences from genuinely random states when assessed using this metric, which quantifies the amount of quantum correlation. Considering initial conditions is therefore vital when interpreting results from quantum simulations and benchmarking experiments. The entanglement entropy serves as a sensitive probe of the system’s state, revealing the extent to which it has deviated from true randomness. Future research could explore methods for quantifying the impact of initial state limitations on the accuracy of quantum simulations and benchmarking procedures, providing guidelines for experimentalists to mitigate these effects.

The research demonstrated that limitations on generating truly random quantum states arise from experimental initialisation, not the symmetries within the systems themselves. Initial states with minimal quantum entanglement actively prevent systems from fully realising randomness, even under chaotic conditions. This means that even at long timescales, states remain distinguishable from genuinely random ones when measured using entanglement entropy. The findings highlight the importance of considering the entire experimental setup, including initial state preparation, when interpreting results from quantum simulations and benchmarking experiments.