Search

Interfacing light from solid-state single-photon sources with scalable and robust room-temperature quantum memories has been a long-standing challenge in photonic quantum information technologies due to inherent noise processes and time-scale mismatches between the operating conditions of solid-state and atomic systems. Here, we demonstrate on-demand storage and retrieval of single photons from a semiconductor quantum dot device in a room-temperature atomic vapor memory. A deterministically fabricated InGaAs quantum dot light source emits single photons at the wavelength of the cesium D1 line at 895\,nm which exhibit an inhomogeneously broadened linewidth of 5.1(7)\,GHz and are subsequently stored in a low-noise ladder-type cesium vapor memory. We show control over the interaction between the single photons and the atomic vapor, allowing for variable retrieval times of up to 19.8(3)\,ns at an internal efficiency of $\eta_\mathrm{int}=0.6(1)\%$. Our results significantly expand the application space of both room-temperature vapor memories and semiconductor quantum dots in future quantum network architectures.

In an era where digital security is crucial, efficient processing of security-related inquiries through supply chain security questionnaires is imperative. This paper introduces a novel approach using Natural Language Processing (NLP) and Retrieval-Augmented Generation (RAG) to automate these responses. We developed QuestSecure, a system that interprets diverse document formats and generates precise responses by integrating large language models (LLMs) with an advanced retrieval system. Our experiments show that QuestSecure significantly improves response accuracy and operational efficiency. By employing advanced NLP techniques and tailored retrieval mechanisms, the system consistently produces contextually relevant and semantically rich responses, reducing cognitive load on security teams and minimizing potential errors. This research offers promising avenues for automating complex security management tasks, enhancing organizational security processes.

Anomaly detection in clinical time-series holds significant potential in identifying suspicious patterns in different biological parameters. In this paper, we propose a targeted method that incorporates the clinical domain knowledge into LLMs to improve their ability to detect anomalies. We introduce the Metabolism Pathway-driven Prompting (MPP) method, which integrates the information about metabolic pathways to better capture the structural and temporal changes in biological samples. We applied our method for doping detection in sports, focusing on steroid metabolism, and evaluated using real-world data from athletes. The results show that our method improves anomaly detection performance by leveraging metabolic context, providing a more nuanced and accurate prediction of suspicious samples in athletes' profiles.

Magnetotaxis is a well known phenomenon in swimming microorganisms which sense magnetic fields e.g. by incorporating crystalline magnetosomes. In designing artificial active matter with tunable dynamics, external magnetic fields can provide a versatile method for guidance. Here, it is the question what material properties are necessary to elicit a significant response. In this working paper, we document in experiments on self-propelling nematic microdroplets that the weak diamagnetic torques exerted by a sub-Tesla magnetic field are already sufficient to significantly affect the dynamics of these swimmers. We find a rich and nontrivial variety of dynamic modes by varying droplet size, state of confinement and magnetic field strength.

Rodents employ a broad spectrum of ultrasonic vocalizations (USVs) for social communication. As these vocalizations offer valuable insights into affective states, social interactions, and developmental stages of animals, various deep learning approaches have aimed to automate both the quantitative (detection) and qualitative (classification) analysis of USVs. Here, we present the first systematic evaluation of different types of neural networks for USV classification. We assessed various feedforward networks, including a custom-built, fully-connected network and convolutional neural network, different residual neural networks (ResNets), an EfficientNet, and a Vision Transformer (ViT). Paired with a refined, entropy-based detection algorithm (achieving recall of 94.9% and precision of 99.3%), the best architecture (achieving 86.79% accuracy) was integrated into a fully automated pipeline capable of analyzing extensive USV datasets with high reliability. Additionally, users can specify an individual minimum accuracy threshold based on their research needs. In this semi-automated setup, the pipeline selectively classifies calls with high pseudo-probability, leaving the rest for manual inspection. Our study focuses exclusively on neonatal USVs. As part of an ongoing phenotyping study, our pipeline has proven to be a valuable tool for identifying key differences in USVs produced by mice with autism-like behaviors.

Ferroic materials enable a multitude of emerging applications, and optimum functional properties are achieved when ferromagnetic and ferroelectric properties are coupled to a first-order ferroelastic transition. In bulk materials, this first-order transition involves an invariant habit plane, connecting coexisting phases: austenite and martensite. Theory predicts that this plane should converge to a line in thin films, but experimental evidence is missing. Here, we analyze the martensitic and magnetic microstructure of a freestanding epitaxial magnetic shape memory film. We show that the martensite microstructure is determined by an invariant line constraint using lattice parameters of both phases as the only input. This line constraint explains most of the observable features, which differ fundamentally from bulk and constrained films. Furthermore, this finite-size effect creates a remarkable checkerboard magnetic domain pattern through multiferroic coupling. Our findings highlight the decisive role of finite-size effects in multiferroics., Comment: 27 pages, 6 figures

We determine distributions and correlation properties of offshore wind speeds and wind speed increments by analyzing wind data sampled with a resolution of one second for 20 months at different heights above sea level in the North Sea. Distributions of horizontal wind speeds can be fitted to Weibull distributions with shape and scale parameters varying weakly with the vertical height separation. Kullback-Leibler divergences between distributions at different heights change with the squared logarithm of the height ratio. Cross-correlations between time derivatives of wind speeds are long-term anticorrelated, and the even parts of their correlation functions satisfy sum rules. Distributions of horizontal wind speed increments change from a tent-like shape to a Gaussian with rising increment lag. A surprising peak occurs in the left tail of the increment distributions for lags in a range $10-200\,{\rm km}$ after applying the Taylor's hypothesis locally to transform time lags into distances. The peak is decisive in order to obtain an expected and observed linear scaling of third-order structure functions with distance. This suggests that it is an intrinsic feature of atmospheric turbulence., Comment: 20 pages, 14 figures

This paper presents the concept of an ultra-stable, thermally independent precision voltage divider tailored for direct current (DC) voltages up to 60 kV. Key features of this voltage divider include minimal voltage dependence, excellent stability, and resistance to external temperature variations. The innovative approach involves its fabrication using 3D printing technology, allowing easy replication by project partners. This precision voltage divider leverages commercially available precision resistors, drawing upon successful outcomes from the FutureEnergy 19ENG02 and HVDC ENG07 Projects. In these experiments, which involve ion acceleration and laser probing of electronic transitions, voltage dividers are integrated into setups such as COALA (TU Darmstadt), BECOLA (Michigan State University), COLLAPS (CERN/ISOLDE), and ATLANTIS (Argonne National Laboratory). Monitoring the applied acceleration potential, these dividers allow one to consider and counteract long-term drifts and thereby improving measurement accuracy.

We present an efficient method to perform overdamped Brownian dynamics simulations in external force fields and for particle interactions that include a hardcore part. The method applies to particle motion in one dimension, where it is possible to update particle positions by repositioning particle clusters as a whole. These clusters consist of several particles in contact. They form because particle collisions are treated as completely inelastic rather than elastic ones. Updating of cluster positions in time steps is carried out by cluster fragmentation and merging procedures. The presented method is particularly powerful at high collision rates in densely crowded systems, where collective movements of particle assemblies is governing the dynamics. As an application, we simulate the single-file diffusion of sticky hard spheres in a periodic potential., Comment: 15 pages, 7 figures

We develop a Mean-Field (MF) view of the learning dynamics of overparametrized Artificial Neural Networks (NN) under data symmetric in law wrt the action of a general compact group $G$. We consider for this a class of generalized shallow NNs given by an ensemble of $N$ multi-layer units, jointly trained using stochastic gradient descent (SGD) and possibly symmetry-leveraging (SL) techniques, such as Data Augmentation (DA), Feature Averaging (FA) or Equivariant Architectures (EA). We introduce the notions of weakly and strongly invariant laws (WI and SI) on the parameter space of each single unit, corresponding, respectively, to $G$-invariant distributions, and to distributions supported on parameters fixed by the group action (which encode EA). This allows us to define symmetric models compatible with taking $N\to\infty$ and give an interpretation of the asymptotic dynamics of DA, FA and EA in terms of Wasserstein Gradient Flows describing their MF limits. When activations respect the group action, we show that, for symmetric data, DA, FA and freely-trained models obey the exact same MF dynamic, which stays in the space of WI laws and minimizes therein the population risk. We also give a counterexample to the general attainability of an optimum over SI laws. Despite this, quite remarkably, we show that the set of SI laws is also preserved by the MF dynamics even when freely trained. This sharply contrasts the finite-$N$ setting, in which EAs are generally not preserved by unconstrained SGD. We illustrate the validity of our findings as $N$ gets larger in a teacher-student experimental setting, training a student NN to learn from a WI, SI or arbitrary teacher model through various SL schemes. We last deduce a data-driven heuristic to discover the largest subspace of parameters supporting SI distributions for a problem, that could be used for designing EA with minimal generalization error.

We investigate how generated structures of GANs correlate with their activations in hidden layers, with the purpose of better understanding the inner workings of those models and being able to paint structures with unconditionally trained GANs. This gives us more control over the generated images, allowing to generate them from a semantic segmentation map while not requiring such a segmentation in the training data. To this end we introduce the concept of tileable features, allowing us to identify activations that work well for painting.

Electrical impedance tomography (EIT) plays a crucial role in non-invasive imaging, with both medical and industrial applications. In this paper, we present three data-driven reconstruction methods for EIT imaging. These three approaches were originally submitted to the Kuopio tomography challenge 2023 (KTC2023). First, we introduce a post-processing approach, which achieved first place at KTC2023. Further, we present a fully learned and a conditional diffusion approach. All three methods are based on a similar neural network as a backbone and were trained using a synthetically generated data set, providing with an opportunity for a fair comparison of these different data-driven reconstruction methods.

In this work, we investigate methods to reduce the noise in deep saliency maps coming from convolutional downsampling. Those methods make the investigated models more interpretable for gradient-based saliency maps, computed in hidden layers. We evaluate the faithfulness of those methods using insertion and deletion metrics, finding that saliency maps computed in hidden layers perform better compared to both the input layer and GradCAM. We test our approach on different models trained for image classification on ImageNet1K, and models trained for tumor detection on Camelyon16 and in-house real-world digital pathology scans of stained tissue samples. Our results show that the checkerboard noise in the gradient gets reduced, resulting in smoother and therefore easier to interpret saliency maps.

In this paper, we study (homogeneous) predecessor-following platoons in which the vehicle-to-vehicle (V2V) communications are affected by random packet losses. We model the overall platoon as a stochastic hybrid system and analyse its string stability via a small-gain approach. For nonlinear platoons, we illustrate how the different elements of the platoon have an impact on string stability, such as platoon topology and vehicle scheduling. For linear time-invariant platoons, we provide an explicit string stability condition that illustrates the interplay between the channel success probability, transmission rate, and time headway constant. Lastly, we illustrate our results by numerical simulations., Comment: 13 pages, 8 figures

N-heterotriangulenes (N-HTAs) represent a class of functional molecules with high potential for optoelectronic materials, for example as electron donating compounds in donor/acceptor (D/A) systems. The capability of two different N-HTAs, N-HTA 550 and N-HTA 557, the latter containing an additional 7-membered ring, to act as electron donors at interfaces with strong tetracyanoquinodimethane (TCNQ and F4TCNQ) acceptors is studied using high-resolution electron energy loss spectroscopy in combination with state-of-the-art quantum chemical calculations. For TCNQ/N-HTA bilayer systems adsorbed on Au(111) Low-energy (< 2.5 eV) electronic transitions which are attributed to charge transfer (CT) states for all four D/A combinations are identified. Based on substantial quantum chemical calculations a generation of ground state CT complexes is excluded. Instead, CT in the excited state, in which an electron-stimulated CT from the N-HTAs to TCNQs is the underlying process, is proposed. The energies of the CT states are determined by the values of the ionization potential and electron affinity of the involved donor and acceptor., Comment: 22 pages, 6 figures, 3 tables

This paper addresses the string stabilization of vehicular platooning when stochastic phenomena are inherent in inter-vehicle communication. To achieve this, we first provide two definitions to analytically assess the string stability in stochastic scenarios, considering the mean and variance of tracking errors as the platoon size grows. Subsequently, we analytically derive necessary and sufficient conditions to achieve this notion of string stability in predecessor-following linear platoons that communicate through additive white noise channels. We conclude that the condition ensuring string stability with ideal communication is essentially the same that achieves stochastic string stability when additive noise channels are in place and guarantees that the tracking error means and variances converge., Comment: 12 pages, 6 figures, preprint submitted to Automatica

Magnetic particle imaging is a tracer-based tomographic imaging technique that allows the concentration of magnetic nanoparticles to be determined with high spatio-temporal resolution. To reconstruct an image of the tracer concentration, the magnetization dynamics of the particles must be accurately modeled. A popular ensemble model is based on solving the Fokker-Plank equation, taking into account either Brownian or N\'eel dynamics. The disadvantage of this model is that it is computationally expensive due to an underlying stiff differential equation. A simplified model is the equilibrium model, which can be evaluated directly but in most relevant cases it suffers from a non-negligible modeling error. In the present work, we investigate an extended version of the equilibrium model that can account for particle anisotropy. We show that this model can be expressed as a series of Bessel functions, which can be truncated based on a predefined accuracy, leading to very short computation times, which are about three orders of magnitude lower than equivalent Fokker-Planck computation times. We investigate the accuracy of the model for 2D Lissajous magnetic particle imaging sequences and show that the difference between the Fokker-Planck and the equilibrium model with anisotropy is sufficiently small so that the latter model can be used for image reconstruction on experimental data with only marginal loss of image quality, even compared to a system matrix-based reconstruction., Comment: This work has been submitted to the IEEE for possible publication

Transport processes in crowded periodic structures are often mediated by cooperative movements of particles forming clusters. Recent theoretical and experimental studies of driven Brownian motion of hard spheres showed that cluster-mediated transport in one-dimensional periodic potentials can proceed in form of solitary waves. We here give a comprehensive description of these solitons. Fundamental for our analysis is a static presoliton state, which is formed by a periodic arrangement of basic stable clusters. Their size follows from a geometric principle of minimum free space. Adding one particle to the presoliton state gives rise to solitons. We derive the minimal number of particles needed for soliton formation, number of solitons at larger particle numbers, soliton velocities and soliton-mediated particle currents. Incomplete relaxations of the basic clusters are responsible for an effective repulsive soliton-soliton interaction seen in measurements. A dynamical phase transition is predicted to occur in current-density relations at low temperatures. Our results provide a theoretical basis for describing experiments on cluster-mediated particle transport in periodic potentials., Comment: 21 pages, 12 figures

We study the topological recurrence phenomenon of actions of locally compact abelian groups on compact metric spaces. In the case of $\mathbb{Z}^d$-actions we develop new techniques to analyze Bohr recurrence sets. These techniques include finding and manipulating correlations between the coordinates of the set of recurrence. Using this, we show that Bohr recurrence sets are recurrence sets for $\mathbb{Z}^d$-Weyl systems. This family encompasses, for example, all $\mathbb{Z}^d$-affine nilsystems. To our knowledge, this is the first result towards a positive answer to Katznelson question in the case of $\mathbb{Z}^d$-actions., Comment: Comments welcome!

On-demand storage and retrieval of quantum information in coherent light-matter interfaces is a key requirement for future quantum networking and quantum communication applications. Alkali vapor memories offer scalable and robust high-bandwidth storage at high repetition rates which makes them a natural fit to interface with solid-state single-photon sources. Here, we experimentally realize a room-temperature ladder-type atomic vapor memory that operates on the Cs D1 line. We provide a detailed experimental characterization and demonstration of on-demand storage and retrieval of weak coherent laser pulses with 0.06 photons per pulse at a high signal-to-noise ratio of SNR$=830(80)$. The memory achieves a maximum internal storage efficiency of $\eta_{\text{int}}=15(1)\%$ and an estimated $1/e$-storage time of $\tau_{\mathrm{s}}\approx32\,$ns. Benchmark properties for the storage of single photons from inhomogeneously broadened state-of-the-art solid-state emitters are estimated from the performance of the memory. Together with the immediate availability of high-quality InGaAs quantum dots emitting at 895\,nm, these results provide clear prospects for the development of a heterogeneous on-demand quantum light interface.

As humans advance toward a higher level of artificial intelligence, it is always at the cost of escalating computational resource consumption, which requires developing novel solutions to meet the exponential growth of AI computing demand. Neuromorphic hardware takes inspiration from how the brain processes information and promises energy-efficient computing of AI workloads. Despite its potential, neuromorphic hardware has not found its way into commercial AI data centers. In this article, we try to analyze the underlying reasons for this and derive requirements and guidelines to promote neuromorphic systems for efficient and sustainable cloud computing: We first review currently available neuromorphic hardware systems and collect examples where neuromorphic solutions excel conventional AI processing on CPUs and GPUs. Next, we identify applications, models and algorithms which are commonly deployed in AI data centers as further directions for neuromorphic algorithms research. Last, we derive requirements and best practices for the hardware and software integration of neuromorphic systems into data centers. With this article, we hope to increase awareness of the challenges of integrating neuromorphic hardware into data centers and to guide the community to enable sustainable and energy-efficient AI at scale., Comment: 11 pages, 2 figures, presented as poster at NICE 2024, 2nd version with updated author list and minor updates

Ciliated microswimmers and flagellated bacteria alter their swimming trajectories to follow the direction of an applied electric field exhibiting electrotaxis. Both for matters of application and physical modelling, it is instructive to study such behaviour in synthetic swimmers. We show here that under an external electric field, self-propelling active droplets autonomously modify their swimming trajectories in microchannels, even undergoing `U-turns', to exhibit robust electrotaxis. Depending on the relative initial orientations of the microswimmer and the external electric field, the active droplet can also navigate upstream of an external flow following a centre-line motion, instead of the oscillatory upstream trajectory observed in absence of electric field. Using a hydrodynamic theory model, we show that the electrically induced angular velocity and electrophoretic effects, along with the microswimmer motility and its hydrodynamic interactions with the microchannel walls, play crucial roles in dictating the electrotactic trajectories and dynamics. Specifically, the transformation in the trajectories during upstream swimming against an external flow under an electric field can be understood as a reverse Hopf bifurcation for a dynamical system. Our study provides a simple methodology and a systematic understanding of manoeuvring active droplets in microconfinements for micro-robotic applications especially in biotechnology.

Quantum computing (QC) and machine learning (ML), taken individually or combined into quantum-assisted ML (QML), are ascending computing paradigms whose calculations come with huge potential for speedup, increase in precision, and resource reductions. Likely improvements for numerical simulations in engineering imply the possibility of a strong economic impact on the manufacturing industry. In this project report, we propose a framework for a quantum computing-enhanced service ecosystem for simulation in manufacturing, consisting of various layers ranging from hardware to algorithms to service and organizational layers. In addition, we give insight into the current state of the art of applications research based on QC and QML, both from a scientific and an industrial point of view. We further analyse two high-value use cases with the aim of a quantitative evaluation of these new computing paradigms for industrially-relevant settings., Comment: 11 pages, 3 figures. Accepted version

Recommendation systems are ubiquitous, from Spotify playlist suggestions to Amazon product suggestions. Nevertheless, depending on the methodology or the dataset, these systems typically fail to capture user preferences and generate general recommendations. Recent advancements in Large Language Models (LLM) offer promising results for analyzing user queries. However, employing these models to capture user preferences and efficiency remains an open question. In this paper, we propose LLMRS, an LLM-based zero-shot recommender system where we employ pre-trained LLM to encode user reviews into a review score and generate user-tailored recommendations. We experimented with LLMRS on a real-world dataset, the Amazon product reviews, for software purchase use cases. The results show that LLMRS outperforms the ranking-based baseline model while successfully capturing meaningful information from product reviews, thereby providing more reliable recommendations.

We develop combinatorial tools to study partial rigidity within the class of minimal $\mathcal{S}$-adic subshifts. By leveraging the combinatorial data of well-chosen Kakutani-Rokhlin partitions, we establish a necessary and sufficient condition for partial rigidity. Additionally, we provide an explicit expression to compute the partial rigidity rate and an associated partial rigidity sequence. As applications, we compute the partial rigidity rate for a variety of constant length substitution subshifts, such as the Thue-Morse subshift, where we determine a partial rigidity rate of 2/3. We also exhibit non-rigid substitution subshifts with partial rigidity rates arbitrarily close to 1 and as a consequence, using products of the aforementioned substitutions, we obtain that any number in $[0, 1]$ is the partial rigidity rate of a system., Comment: Comments welcome!