Publications

The experimental realization of increasingly complex synthetic quantum systems calls for the development of general theoretical methods to validate and fully exploit quantum resources. Quantum state tomography (QST) aims to reconstruct the full quantum state from simple measurements, and therefore provides a key tool to obtain reliable analytics. However, exact brute-force approaches to QST place a high demand on computational resources, making them unfeasible for anything except small systems. Here we show how machine learning techniques can be used to perform QST of highly entangled states with more than a hundred qubits, to a high degree of accuracy. We demonstrate that machine learning allows one to reconstruct traditionally challenging many-body quantities—such as the entanglement entropy—from simple, experimentally accessible measurements. This approach can benefit existing and future generations of devices ranging from quantum computers to ultracold-atom quantum simulators.

We develop a constructive approach to generate artificial neural networks representing the exact ground states of a large class of many-body lattice Hamiltonians. It is based on the deep Boltzmann machine architecture, in which two layers of hidden neurons mediate quantum correlations among physical degrees of freedom in the visible layer. The approach reproduces the exact imaginary-time Hamiltonian evolution, and is completely deterministic. In turn, compact and exact network representations for the ground states are obtained without stochastic optimization of the network parameters. The number of neurons grows linearly with the system size and total imaginary time, respectively. Physical quantities can be measured by sampling configurations of both physical and neuron degrees of freedom. We provide specific examples for the transverse-field Ising and Heisenberg models by implementing efficient sampling. As a compact, classical representation for many-body quantum systems, our approach is an alternative to the standard path integral, and it is potentially useful also to systematically improve on numerical approaches based on the restricted Boltzmann machine architecture.

Laser control of solids has so far mainly been discussed in the context of strong classical nonlinear light-matter coupling in a pump-probe framework. Here we propose a quantum-electrodynamical setting to address the coupling of a low-dimensional quantum material to quantized electromagnetic fields in nanocavities. Using a protoypical model system describing FeSe/SrTiO3, we study how the formation of phonon polaritons at the 2D interface of the material modifies its superconducting properties in a Migdal-Eliashberg simulation. We find that through highly polarizable dipolar phonons, cavity-induced superconductivity is possible at temperatures above the bare critical temperature of the system. Our results demonstrate that quantum cavities enable the engineering of fundamental couplings in solids paving the way to unprecedented control of material properties.

We show using theory and experiments that a small particle moving along an elastic membrane through a viscous fluid is repelled from the membrane due to hydro-elastic forces. The viscous stress field produces an elastic disturbance leading to particle-wave coupling. We derive an analytic expression for the particle trajectory in the lubrication limit, bypassing the construction of the detailed velocity and pressure fields. The normal force is quadratic in the parallel speed, and is a function of the tension and bending resistance of the membrane. Experimentally, we measure the normal displacement of spheres sedimenting along an elastic membrane and find quantitative agreement with the theoretical predictions with no fitting parameters. We experimentally demonstrate the effect to be strong enough for particle separation and sorting. We discuss the significance of these results for bio-membranes and propose our model for membrane elasticity measurements.

We develop a first-principles simulation method for attosecond time-resolved photoelectron spectroscopy. This method enables us to directly simulate the whole experimental processes, including excitation, emission and detection on equal footing. To examine the performance of the method, we use it to compute the reconstruction of attosecond beating by interference of two-photon transitions (RABBITT) experiments of gas-phase Argon. The computed RABBITT photoionization delay is in very good agreement with recent experimental results from [Kl\"under et al, Phys. Rev. Lett. 106 143002 (2011)] and [Gu\'enot et al, Phys. Rev. A 85 053424 (2012)]. This indicates the significance of a fully-consistent theoretical treatment of the whole measurement process to properly describe experimental observables in attosecond photoelectron spectroscopy. The present framework opens the path to unravel the microscopic processes underlying RABBITT spectra in more complex materials and nanostructures.

Cilia and flagella are essential building blocks for biological fluid transport and locomotion at the micrometre scale. They often beat in synchrony and may transition between different synchronization modes in the same cell type. Here, we investigate the behaviour of elastic microfilaments, protruding from a surface and driven at their base by a configuration-dependent torque. We consider full hydrodynamic interactions among and within filaments and no slip at the surface. Isolated filaments exhibit periodic deformations, with increasing waviness and frequency as the magnitude of the driving torque increases. Two nearby but independently driven filaments synchronize their beating in-phase or anti-phase. This synchrony arises autonomously via the interplay between hydrodynamic coupling and filament elasticity. Importantly, in-phase and anti-phase synchronization modes are bistable and coexist for a range of driving torques and separation distances. These findings are consistent with experimental observations of in-phase and anti-phase synchronization in pairs of cilia and flagella and could have important implications on understanding the biophysical mechanisms underlying transitions between multiple synchronization modes.

One-dimensional crystals of passively-driven particles in microfluidic channels exhibit collective vibrational modes reminiscent of acoustic ‘phonons’. These phonons are induced by the long-range hydrodynamic interactions among the particles and are neutrally stable at the linear level. Here, we analyze the effect of particle activity – self-propulsion – on the emergence and stability of these phonons. We show that the direction of wave propagation in active crystals is sensitive to the intensity of the background flow. We also show that activity couples, at the linear level, transverse waves to the particles' rotational motion, inducing a new mode of instability that persists in the limit of large background flow, or, equivalently, vanishingly small activity. We then report a new phenomenon of phonons switching back and forth between two adjacent crystals in both passively-driven and active systems, similar in nature to the wave switching observed in quantum mechanics, optical communication, and density stratified fluids. These findings could have implications for the design of commercial microfluidic systems and the self-assembly of passive and active micro-particles into one-dimensional structures.

Observational tests of stellar and Galactic chemical evolution call for the joint knowledge of a star's physical parameters, detailed element abundances, and precise age. For cool main-sequence (MS) stars the abundances of many elements can be measured from spectroscopy, but ages are very hard to determine. The situation is different if the MS star has a white dwarf (WD) companion and a known distance, as the age of such a binary system can then be determined precisely from the photometric properties of the cooling WD. As a pilot study for obtaining precise age determinations of field MS stars, we identify nearly one hundred candidates for such wide binary systems: a faint WD whose GPS1 proper motion matches that of a brighter MS star in Gaia/TGAS with a good parallax (σϖ/ϖ≤0.05). We model the WD's multi-band photometry with the BASE-9 code using this precise distance (assumed to be common for the pair) and infer ages for each binary system. The resulting age estimates are precise to ≤10% (≤20%) for 42 (67) MS-WD systems. Our analysis more than doubles the number of MS-WD systems with precise distances known to date, and it boosts the number of such systems with precise age determination by an order of magnitude. With the advent of the Gaia DR2 data, this approach will be applicable to a far larger sample, providing ages for many MS stars (that can yield detailed abundances for over 20 elements), especially in the age range 2 to 8\,\Gyr, where there are only few known star clusters.

The generation of high-order harmonics from atomic and molecular gases enables the production of high-energy photons and ultrashort isolated pulses. Obtaining efficiently similar photon energy from solid-state systems could lead, for instance, to more compact extreme ultraviolet and soft x-ray sources. We demonstrate from ab initio si- mulations that it is possible to generate high-order harmonics from free-standing monolayer materials, with an energy cutoff similar to that of atomic and molecular gases. In the limit in which electrons are driven by the pump laser perpendicularly to the monolayer, they behave qualitatively the same as the electrons responsible for high- harmonic generation (HHG) in atoms, where their trajectories are described by the widely used semiclassical model, and exhibit real-space trajectories similar to those of the atomic case. Despite the similarities, the first and last steps of the well-established three-step model for atomic HHG are remarkably different in the two-dimensional materials from gases. Moreover, we show that the electron-electron interaction plays an important role in harmonic gener- ation from monolayer materials because of strong local-field effects, which modify how the material is ionized. The recombination of the accelerated electron wave packet is also found to be modified because of the infinite extension of the material in the monolayer plane, thus leading to a more favorable wavelength scaling of the harmonic yield than in atomic HHG. Our results establish a novel and efficient way of generating high-order harmonics based on a solid-state device, with an energy cutoff and a more favorable wavelength scaling of the harmonic yield similar to those of atomic and molecular gases. Two-dimensional materials offer a unique platform where both bulk and atomic HHG can be investigated, depending on the angle of incidence. Devices based on two-dimensional materials can extend the limit of existing sources.

Estimates of the Hubble constant, H0, from the distance ladder and the cosmic microwave background (CMB) differ at the ∼3-σ level, indicating a potential issue with the standard ΛCDM cosmology. Interpreting this tension correctly requires a model comparison calculation depending on not only the traditional `n-σ' mismatch but also the tails of the likelihoods. Determining the form of the tails of the local H0 likelihood is impossible with the standard Gaussian least-squares approximation, as it requires using non-Gaussian distributions to faithfully represent anchor likelihoods and model outliers in the Cepheid and supernova (SN) populations, and simultaneous fitting of the full distance-ladder dataset to correctly propagate uncertainties. We have developed a Bayesian hierarchical model that describes the full distance ladder, from nearby geometric anchors through Cepheids to Hubble-Flow SNe. This model does not rely on any distributions being Gaussian, allowing outliers to be modeled and obviating the need for arbitrary data cuts. Sampling from the ∼3000-parameter joint posterior using Hamiltonian Monte Carlo, we find H0 = (72.72 ± 1.67) kms−1Mpc−1 when applied to the outlier-cleaned Riess et al. (2016) data, and (73.15±1.78) kms−1Mpc−1 with SN outliers reintroduced. Our high-fidelity sampling of the low-H0 tail of the distance-ladder likelihood allows us to apply Bayesian model comparison to assess the evidence for deviation from ΛCDM. We set up this comparison to yield a lower limit on the odds of the underlying model being ΛCDM given the distance-ladder and Planck XIII (2016) CMB data. The odds against ΛCDM are at worst 10:1 or 7:1, depending on whether the SNe outliers are cut or modeled, or 60:1 if an approximation to the Planck Int. XLVI (2016) likelihood is used.