Squeezed states, squashed states, spoofing, and more

Rationalizing systematic discrepancies between election outcomes and opinion polls

This work analyzes an exactly-solvable Ising model of the Bradley effect, which refers to discrepancies between election results and opinion polls resulting from poll respondents hiding their true preference. The Ising model takes the form of a bipartite network composed of a hidden layer (corresponding to voter intentions) and a visible layer (the voters' declared preferences).

Quadrature nonreciprocity: unidirectional bosonic transmission without breaking time-reversal symmetry

The authors propose and experimentally demonstrate quadrature non-reciprocity, referring to unidirectional transport of mode quadratures arising due to the interplay between linear coupling (beam splitter operations) and two-mode squeezing. This platform promises to be an interesting playground for non-Hermitian physics, with potential applications for multi-mode bosonic quantum networks. 

When you can't count, sample! Computable entropies beyond equilibrium from basin volumes

A very clear and beginner-friendly perspective on the difficulty of estimating volumes and densities in high-dimensional configuration spaces, and how the problem can be efficiently solved using importance sampling and replicas. Verification of quantum supremacy experiments is tricky for similar reasons - estimation of the probability density function of a random quantum state requires an exponential number of measurements.

Fundamental constraints on the observability of non-Hermitian effects in passive systems

There is still huge interest in non-Hermitian phenomena including non-Hermitian topological phases. Authors on this subject would be wise to understand important caveats regarding commonly-used models. For example, when considering a model including gain terms, some kind of nonlinear term such as gain saturation is required to avoid an unphysical blow-up of energy. Analysis of nonlinear systems is of course much more difficult, making it hard to obtain rigorous general results in this case. One alternative is to consider purely passive non-Hermitian systems formed via inhomogeneous losses, such that one does not need to worry about the stability of the system. Interestingly, this is not the whole story. This preprint shows that when inevitable material dispersion is included into passive non-Hermitian Hamiltonians (which is required for the model to satisfy the fundamental constraint of causality), "some of the most widely studied features [exceptional points, non-Hermitian skin effect, and symmetry-protected edge states] are effectively disguised in the density of states, in particular to the signatures of drastic mode nonorthogonality. These findings highlight the essential role of active elements in devices that aim to exploit these signatures."

Solving the sampling problem of the Sycamore quantum supremacy circuits 

This preprint from last year was just accepted in PRL. The authors demonstrate a method to classically spoof the results of the Google quantum supremacy experiments. "If our algorithm could be implemented with high efficiency on a modern supercomputer with ExaFLOPS performance, we estimate that ideally, the simulation would cost a few dozens of seconds, which is faster than Google's quantum hardware."

Classical models are a better explanation of the Jiuzhang Gaussian Boson Samplers than their targeted squeezed light models

Nicolas Quesada (formerly of Xanadu) and collaborators show that mixtures of coherent states termed squashed states are capable of spoofing higher order correlations measured in Gaussian BosonSampling experiments, one of the (multiple) tests used to establish a potential quantum advantage. However, the results of another test (Heavy Output Generation), could not be classically reproduced using the squashed state. "This work thus provides a new adversary that should be considered against future GBS experiments and, perhaps more importantly, further motivates the need to identify proper metrics and optimal classical adversaries for quantum advantage in the context of threshold GBS"