Establishing a link between statistical properties and genuine randomness in quantum processes proved elusive until now. Matching statistical moments, specifically creating approximate unitary designs, does not guarantee pseudorandomness in quantum systems. Understanding how simple processes can exhibit complex behaviour has been a longstanding goal.

Recent work investigates this question within quantum systems, building on the classical idea that matching statistical patterns should guarantee unpredictability, much like testing whether shuffling cards thoroughly produces an unbiased deck. However, replicating certain statistical properties, specifically through unitary designs, a way of mathematically checking if a series of quantum operations appears random by examining its average behaviour across multiple measurements, does not automatically create genuine randomness or ‘pseudorandomness’, the quality of appearing unpredictable despite being generated deterministically.

This finding challenges a long-held idea, similar to a belief about permutations, which suggested mathematical independence guarantees unpredictability. Unitary designs, used to approximate complex behaviours, are distinguishable from truly random processes using an efficient algorithm after a specific number of steps. The team demonstrated this with an algorithm capable of distinguishing these imperfectly random ensembles from truly random ones in Ot(n² log² n) steps; akin to saying building a tower takes longer and requires more bricks as you want it taller. Matching statistical patterns alone does not create genuine randomness within quantum systems.

Efficient construction of low complexity approximate unitary t-designs

The work centred on constructing and analysing specific types of quantum operations known as approximate unitary t-designs, a mathematical tool used to assess how closely a series of quantum steps mimics truly random behaviour by examining its average effects across many measurements. An ensemble of gates, combinations of actions on single and pairs of qubits, was engineered for rapid construction requiring Ot(n² log² n) computational steps; this parallels the increasing effort needed when building a taller tower with more bricks.

Unitary designs, ensembles matching initial statistical properties of genuinely random processes, were investigated alongside their relationship to creating unpredictable systems. This process scaled similarly to Ot(n² log² n), allowing efficient sampling of gate distributions while alternatives often require sharply more complex calculations or lack control over ensemble structure when modelling quantum behaviour.

Statistical Moments Do Not Imply Quantum Pseudorandomness Demonstrated Through Efficient Ensemble Construction

This represents a strong improvement on previous work as it establishes a clear separation between matching statistical moments and achieving true pseudorandomness, something previously thought to be intrinsically linked. Prior research required sharply more computational effort than now proven possible when creating an ensemble statistically indistinguishable from random while remaining distinguishable itself. The team refuted the quantum equivalent of a longstanding conjecture concerning permutations by efficiently constructing such an ensemble.

Efficient algorithms can differentiate these ensembles from truly random ones using limited queries; this proves that matching statistical moments in unitary designs does not necessarily imply pseudorandomness, even for simple quantum processes involving one and two-qubit gates. This refutes the quantum analogue of a conjecture proposed by Hoory, Magen, Myers & Rackoff regarding permutations. A stronger separation between unitary designs and pseudorandom unitaries exists at polynomially bounded moments but requires structured arrangements rather than straightforward compositions.

Caution is therefore warranted when applying unitary designs to modelling information scrambling within black-hole physics as structure may remain accessible via experimentation. Clarifying that these designs do not ensure genuine pseudorandomness defines limitations when applying such approximations to complex systems like those found in theoretical physics.

Despite this subtlety, the research remains valuable because it precisely pinpoints where current methods fall short of true unpredictability and motivates exploration into more robust approaches for generating randomness within quantum processes. Statistical similarity alone does not guarantee genuine unpredictability. This distinction is significant as previous assumptions linked matching mathematical patterns with the emergence of pseudorandomness, an idea mirrored in classical problems involving permutations and shuffling. Scientists have identified limitations when applying these approximations to modelling complex physical phenomena such as information scrambling inside black holes by constructing efficiently samplable gate distributions exhibiting this separation.

The research demonstrated that approximating random quantum behaviours using unitary designs does not necessarily create genuinely unpredictable systems. Specifically, scientists constructed examples utilising one and two-qubit gates where ensembles matched statistical properties but were still identifiable as non-random following a certain number of processing steps. These results refute a previously held conjecture regarding the relationship between matching mathematical patterns and true randomness within quantum processes. This clarifies limits in employing such approximations for modelling complex physics like information scrambling in black holes, because structure can remain detectable even in seemingly randomised systems.

👉 More information
🗞 On the pseudorandomness of simple quantum processes
✍️ Jesko Dujmovic, Jonas Haferkamp and Alexander Poremba (Boston University)
🧠 ArXiv: https://arxiv.org/abs/2610.02100

More like this