Search

We study the inverse problem of determining a time-dependent globally hyperbolic Lorentzian metric from the scattering operator for semilinear wave equations., Comment: 62 pages

Modern imaging technologies are widely based on classical principles of light or electromagnetic wave propagation. They can be remarkably sophisticated, with recent successes ranging from single molecule microscopy to imaging far-distant galaxies. However, new imaging technologies based on quantum principles are gradually emerging. They can either surpass classical approaches or provide novel imaging capabilities that would not otherwise be possible. {Here }we provide an overview {of the most recently developed quantum imaging systems, highlighting the non-classical properties of sources such as bright squeezed light, entangled photons, and single-photon emitters that enable their functionality.} We outline potential upcoming trends and the associated challenges, all driven by a central inquiry, which is to understand whether quantum light can make visible the invisible.

Despite the recent improvements in supporting Persistent Hardware Transactions (PHTs) on emerging persistent memories (PM), the poor performance of Read-Only (RO) transactions remains largely overlooked. We propose DUMBO, a new design for PHT that eliminates the two most crucial bottlenecks that hinder RO transactions in state-of-the-art PHT. At its core, DUMBO exploits advanced instructions that some contemporary HTMs provide to suspend (and resume) transactional access tracking. Our experimental evaluation with an IBM POWER9 system using the TPC-C benchmark shows that DUMBO can outperform the state of the art designs for persistent hardware (SPHT) and software memory transactions (Pisces), by up to 4.0x.

This paper is motivated by studies in neuroscience experiments to understand interactions between nodes in a brain network using different types of data modalities that capture different distinct facets of brain activity. To assess Granger-causality, we introduce a flexible framework through a general class of models that accommodates mixed types of data (binary, count, continuous, and positive components) formulated in a generalized linear model (GLM) fashion. Statistical inference for causality is performed based on both frequentist and Bayesian approaches, with a focus on the latter. Here, we develop a procedure for conducting inference through the proposed Bayesian mixed time series model. By introducing spike and slab priors for some parameters in the model, our inferential approach guides causality order selection and provides proper uncertainty quantification. The proposed methods are then utilized to study the rat spike train and local field potentials (LFP) data recorded during the olfaction working memory task. The proposed methodology provides critical insights into the causal relationship between the rat spiking activity and LFP spectral power. Specifically, power in the LFP beta band is predictive of spiking activity 300 milliseconds later, providing a novel analytical tool for this area of emerging interest in neuroscience and demonstrating its usefulness and flexibility in the study of causality in general., Comment: Paper submitted for publication

We discuss how to extract numerically the expected lowest quasi-normal mode (QNM) associated with the pressure anisotropy for a Bjorken flux evolution to equilibrium. This QNM was easily decoded subtracting the hydrodynamical attractors and was compared with other authors calculations. After evolutions with transients close to the expected lowest QNM the system goes to a tail (pure imaginary frequency) for late times. We analyze the relevance of Navier-Stokes, second order and Borel attractors at each stage of the evolution, which begins far-from-equilibrium and ends close to equilibrium., Comment: 7 pages, 6 figures; revised version; title, abstract and keywords added in Portuguese

We present an experiment investigating the rotational Doppler effect using a single trapped ion excited by two copropagating vortex laser beams. The setup isolates the azimuthal gradients of the fields, eliminating longitudinal and curvature effects. We provide a detailed characterization of the phenomenon by deterministically positioning a single ion across the beams, achieving a signal which depends on the angular velocity of the ion and the difference of optical orbital angular momentum between the two beams. The interpretation of the measurements is supported by numerical simulations and by a simplified analytical model. Our results reveal key properties of the rotational Doppler effect, showing that it increases approaching the center of the beam and that it is independent of the waist of the beam. This offers insights into the feasibility of super-kicks or super-Doppler shifts for sensing and manipulating atomic motion transverse to the beams' propagation direction., Comment: 5+6 pages, 2+4 figures

We study how to optimally segment a monopolistic market given a redistributive objective. Optimal redistributive segmentations (i) induce the seller to price progressively, i.e., richer consumers pay higher prices than poorer ones, and (ii) may require giving a higher profit than uniform pricing if the redistributive motive is strong. We further show that optimal redistributive segmentations are implementable via price-based regulation.

In tissue engineering, 3D scaffolds and chemical treatments are often used for providing a cell-friendly surface for improving cell adhesion and tissue growth. Indeed, the cell adhesion degree can be controlled by physical-chemical changes in the surface of substrates, such as wettability, surface charges and roughness. In this work, we describe the synthesis, characterization and cytocompatibility of photoresins useful for construction of cell scaffolds via two-photon polymerization. Additionally, we have demonstrated a simple surface treatment method that promotes cell adhesion. This method alters the surface charge of the polymer and enhances the adhesion of epithelial cells. Our results indicate an efficient approach for modifying the surface of biocompatible polymer scaffolds with the purpose of enhances the performance of cell functions suitable for tissue engineering and regenerative medicine.

Cosmic rays (CRs) interact with turbulent magnetic fields in the intestellar medium, generating nonthermal emission. After many decades of studies, the theoretical understanding of their diffusion in the ISM continues to pose a challenge. This study numerically explores a recent prediction termed "mirror diffusion" and its synergy with traditional diffusion mechanism based on gyroresonant scattering. Our study combines 3D MHD simulations of star-forming regions with test particle simulations to analyze CR diffusion. We demonstrate the significance of mirror diffusion in CR diffusion parallel to the magnetic field, when the mirroring condition is satisfied. Our results support the theoretical expectation that the resulting particle propagation arising from mirror diffusion in combination with much faster diffusion induced by gyroresonant scattering resembles a Levy-flight-like propagation. Our study highlights the necessity to reevaluate the diffusion coefficients traditionally adopeted in the ISM based on gyroresonant scattering alone. For instance, our simulations imply a diffusion coefficient $\sim10^{27}cm^2/s$ for particles with a few hundred TeV within regions spanning a few parsecs around the source. This estimate is in agreement with gamma-ray observations, which shows the relevance of our results for understanding of diffuse gamma-ray emission in star-forming regions.

Bug fixing is a crucial task in software maintenance to hold user trust. Although various automated fault localization techniques exist, they often require specific conditions to be effective. For example, Spectrum-Based Fault Localization (SBFL) techniques need at least one failing test to identify bugs, which may not always be available. Bug reports, particularly those with stack traces, provide detailed information on system execution failures and are invaluable for developers. This study focuses on utilizing stack traces from crash reports as fault-triggering tests for SBFL. Our findings indicate that only 3.33% of bugs have fault-triggering tests, limiting traditional SBFL efficiency. However, 98.3% of bugfix intentions align directly with exceptions in stack traces, and 78.3% of buggy methods are reachable within an average of 0.34 method calls, proving stack traces as a reliable source for locating bugs. We introduce a new approach, SBEST, that integrates stack trace data with test coverage to enhance fault localization. Our approach shows a significant improvement, increasing Mean Average Precision (MAP) by 32.22% and Mean Reciprocal Rank (MRR) by 17.43% over traditional stack trace ranking methods.

Samples of the monolithic silicon pixel ASIC prototype produced in 2022 within the framework of the Horizon 2020 MONOLITH ERC Advanced project were irradiated with 70 MeV protons up to a fluence of 1 x 1016 neq/cm2, and then tested using a beam of 120 GeV/c pions. The ASIC contains a matrix of 100 \mu m pitch hexagonal pixels, readout out by low noise and very fast frontend electronics produced in a 130 nm SiGe BiCMOS technology process. The dependence on the proton fluence of the efficiency and the time resolution of this prototype was measured with the frontend electronics operated at a power density between 0.13 and 0.9 W/cm2. The testbeam data show that the detection efficiency of 99.96% measured at sensor bias voltage of 200 V before irradiation becomes 96.2% after a fluence of 1 x 1016 neq/cm2. An increase of the sensor bias voltage to 300 V provides an efficiency to 99.7% at that proton fluence. The timing resolution of 20 ps measured before irradiation rises for a proton fluence of 1 x 1016 neq/cm2 to 53 and 45 ps at HV = 200 and 300 V, respectively.

Photonic Integrated Circuits (PIC)s are a promising contender for quantum information technologies. The spectral purity of photons is one of the key attributes of PIC photon-pair sources. The dual-pulse pump manipulation technique [1] showed >99% purity in ring-resonator photon-pair sources. Here, we have developed a PIC to shape a pulse into dual, triple and quadruple pulses and investigated the effect of these pump pulse configurations on the purity. Our results show that more complex configurations over dual-pulse do not result in comparatively higher purity but allow accurate control over choosing arbitrary values of the purity.

Here we show photon pair generation from ring resonator and waveguide structures in a hybrid silicon-BTO on insulator platform with a pulsed pump. Our analysis of single photon and coincidence generation rates show that Spontaneous Four-Wave Mixing is comparable to that expected from SOI devices of similar characteristics and find a $\gamma_{eff}$ of (14.7 $\pm$ 1.3) and (2.0 $\pm$ 0.3) MHz/mW$^{2}$ for ring resonator and waveguide structures respectively.

This paper establishes the optimal sub-Gaussian variance proxy for truncated Gaussian and truncated exponential random variables. The proofs rely on first characterizing the optimal variance proxy as the unique solution to a set of two equations and then observing that for these two truncated distributions, one may find explicit solutions to this set of equations. Moreover, we establish the conditions under which the optimal variance proxy coincides with the variance, thereby characterizing the strict sub-Gaussianity of the truncated random variables. Specifically, we demonstrate that truncated Gaussian variables exhibit strict sub-Gaussian behavior if and only if they are symmetric, meaning their truncation is symmetric with respect to the mean. Conversely, truncated exponential variables are shown to never exhibit strict sub-Gaussian properties. These findings contribute to the understanding of these prevalent probability distributions in statistics and machine learning, providing a valuable foundation for improved and optimal modeling and decision-making processes., Comment: 30 pages, 4 figures

Both text and video data are abundant on the internet and support large-scale self-supervised learning through next token or frame prediction. However, they have not been equally leveraged: language models have had significant real-world impact, whereas video generation has remained largely limited to media entertainment. Yet video data captures important information about the physical world that is difficult to express in language. To address this gap, we discuss an under-appreciated opportunity to extend video generation to solve tasks in the real world. We observe how, akin to language, video can serve as a unified interface that can absorb internet knowledge and represent diverse tasks. Moreover, we demonstrate how, like language models, video generation can serve as planners, agents, compute engines, and environment simulators through techniques such as in-context learning, planning and reinforcement learning. We identify major impact opportunities in domains such as robotics, self-driving, and science, supported by recent work that demonstrates how such advanced capabilities in video generation are plausibly within reach. Lastly, we identify key challenges in video generation that mitigate progress. Addressing these challenges will enable video generation models to demonstrate unique value alongside language models in a wider array of AI applications.

Correlation identities are obtained for $Z_3$ lattice gauge theory where the bonds of the plaquettes are decorated by generalized three-state Ising variables. Making use of correlation inequalities we obtain the area decay of the Wilson loop observable in a range of the coupling parameter larger than those obtained from mean field theory considerations., Comment: 7 pages

This paper contributes a new approach for distributional reinforcement learning which elucidates a clean separation of transition structure and reward in the learning process. Analogous to how the successor representation (SR) describes the expected consequences of behaving according to a given policy, our distributional successor measure (SM) describes the distributional consequences of this behaviour. We formulate the distributional SM as a distribution over distributions and provide theory connecting it with distributional and model-based reinforcement learning. Moreover, we propose an algorithm that learns the distributional SM from data by minimizing a two-level maximum mean discrepancy. Key to our method are a number of algorithmic techniques that are independently valuable for learning generative models of state. As an illustration of the usefulness of the distributional SM, we show that it enables zero-shot risk-sensitive policy evaluation in a way that was not previously possible., Comment: Accepted to ICML 2024. First two authors contributed equally

We compute the time evolution of the non-equilibrium entropy in the homogeneous isotropization dynamics of the 1RCBH model, which has a critical point in its conformal phase diagram defined at finite temperature and R-charge density. We also evaluate the time evolution of the pressure anisotropy and the scalar condensate of the medium. We disclose a new feature (not present in the Bjorken flow dynamics analyzed in previous works), which is observed for all the analyzed initial data: the formation of a periodic sequence of several close plateaus in the form of a stairway for the entropy density near thermodynamic equilibrium. We find that the period of plateau formation in the stairway is half the period of oscillations of the slowest quasinormal mode of the system, which is therefore strongly tied to the late time dissipative dynamics of the system associated to the irreversibility of entropy production. For the particular case of the purely thermal SYM plasma at zero density and vanishing scalar condensate, we find that the period of the stairway is half the period of oscillations of the slowest quasinormal mode associated to the late time equilibration of the pressure anisotropy of the fluid, while at finite chemical potential the slowest quasinormal mode of the system is associated to the late time equilibration of the scalar condensate., Comment: 19 pages, 4 figures, accepted for publication in Physical Review D

We study the influence of micromotion on the spectrum of trapped ions with a lambda-type level scheme, leading to dark resonances due to coherent population trapping. We work with calcium ions trapped in a ring-shaped Paul trap, in which one can compensate excess micromotion for only one ion of the crystal. We observe that micromotion affects the shapes of the dark resonances and causes the appearance of "echoes" separated by intervals given by the drive frequency. We present a theoretical model that provides good fits to the measurements and can be used to estimate the amplitude of the micromotion modulation of the atomic motion. We estimate an effective temperature of the ions from the spectra and observe clear micromotion heating as well as impaired cooling for sufficiently large excess micromotion., Comment: 10 pages, 4 figures

The time resolution of the second monolithic silicon pixel prototype produced for the MONOLITH H2020 ERC Advanced project was studied using a femtosecond laser. The ASIC contains a matrix of hexagonal pixels with 100 {\mu}m pitch, readout by low-noise and very fast SiGe HBT frontend electronics. Silicon wafers with 50 {\mu}m thick epilayer with a resistivity of 350 {\Omega}cm were used to produce a fully depleted sensor. At the highest frontend power density tested of 2.7 W/cm2, the time resolution with the femtosecond laser pulses was found to be 45 ps for signals generated by 1200 electrons, and 3 ps in the case of 11k electrons, which corresponds approximately to 0.4 and 3.5 times the most probable value of the charge generated by a minimum-ionizing particle. The results were compared with testbeam data taken with the same prototype to evaluate the time jitter produced by the fluctuations of the charge collection., Comment: Submitted to JINST

A reliable and cost-effective interconnect technology is required for the development of hybrid pixel detectors. The interconnect technology needs to be adapted for the pitch and die sizes of the respective applications. For small-scale applications and during the ASIC and sensor development phase, interconnect technologies must also be suitable for the assembly of single-dies typically available from Multi-Project-Wafer submissions. Within the CERN EP R&D program and the AIDAinnova collaboration, innovative and scalable hybridization concepts are under development for pixel-detector applications in future colliders. This contribution presents recent results of a newly developed in-house single-die interconnection process based on Anisotropic Conductive Adhesives (ACA). The ACA interconnect technology replaces solder bumps with conductive micro-particles embedded in an epoxy layer applied as either film or paste. The electro-mechanical connection between the sensor and ASIC is achieved via thermocompression of the ACA using a flip-chip device bonder. A specific pixel-pad topology is required to enable the connection via micro-particles and create cavities into which excess epoxy can flow. This pixel-pad topology is achieved with an in-house Electroless Nickel Immersion Gold process that is also under development within the project. The ENIG and ACA processes are qualified with a variety of different ASICs, sensors, and dedicated test structures, with pad diameters ranging from 12 {\mu}m to 140 {\mu}m and pitches between 20 {\mu}m and 1.3 mm. The produced assemblies are characterized electrically, with radioactive-source exposures, and in tests with high-momentum particle beams. A focus is placed on recent optimization of the plating and interconnect processes, resulting in an improved plating uniformity and interconnect yield., Comment: Procceding to IPRD23 conference in Siena 2023

Organic light-emitting transistors (OLETs) are multifunctional optoelectronic devices that combine in a single structure the advantages of organic light emitting diodes (OLEDs) and organic field-effect transistors (OFETs). However, low charge mobility and high threshold voltage are critical hurdles to practical OLETs implementation. This work reports on the improvements obtained by using polyurethane films as dielectric layer material in place of the standard poly(methylmethacrylate) (PMMA) in OLET devices. It was found that polyurethane drastically reduces the number of traps in the device thereby improving electrical and optoelectronic device parameters. In addition, a model was developed to rationalize an anomalous behavior at the pinch-off voltage. Our findings represent a step forward to overcome the limiting factors of OLETs that prevent their use in commercial electronics by providing a simple route for low-bias device operation., Comment: 25 pages, 5 figures, 1 table

In this article, we propose a modified convex combination of the polynomial reconstructions of odd-order WENO schemes to maintain the central substencil prevalence over the lateral ones in all parts of the solution. New "centered" versions of the classical WENO-Z and its less dissipative counterpart, WENO-Z+, are defined through very simple modifications of the classical nonlinear weights and show significantly superior numerical properties; for instance, a well-known dispersion error for long-term runs is fixed, along with decreased dissipation and better shock-capturing abilities. Moreover, the proposed centered version of WENO-Z+ has no ad-hoc parameters and no dependence on the powers of the grid size. All the new schemes are thoroughly analyzed concerning convergence at critical points, adding to the discussion on the relevance of such convergence to the numerical simulation of typical hyperbolic conservation laws problems. Nonlinear spectral analysis confirms the enhancement achieved by the new schemes over the standard ones.

Remote sensing and artificial intelligence are pivotal technologies of precision agriculture nowadays. The efficient retrieval of large-scale field imagery combined with machine learning techniques shows success in various tasks like phenotyping, weeding, cropping, and disease control. This work will introduce a machine learning framework for automatized large-scale plant-specific trait annotation for the use case disease severity scoring for Cercospora Leaf Spot (CLS) in sugar beet. With concepts of Deep Label Distribution Learning (DLDL), special loss functions, and a tailored model architecture, we develop an efficient Vision Transformer based model for disease severity scoring called SugarViT. One novelty in this work is the combination of remote sensing data with environmental parameters of the experimental sites for disease severity prediction. Although the model is evaluated on this special use case, it is held as generic as possible to also be applicable to various image-based classification and regression tasks. With our framework, it is even possible to learn models on multi-objective problems as we show by a pretraining on environmental metadata., Comment: submitted to Computers and Electronics in Agriculture

A monolithic silicon pixel prototype produced for the MONOLITH ERC Advanced project was irradiated with 70 MeV protons up to a fluence of 1 x 10^16 1 MeV n_eq/cm^2. The ASIC contains a matrix of hexagonal pixels with 100 {\mu}m pitch, readout by low-noise and very fast SiGe HBT frontend electronics. Wafers with 50 {\mu}m thick epilayer with a resistivity of 350 {\Omega}cm were used to produce a fully depleted sensor. Laboratory tests conducted with a 90Sr source show that the detector works satisfactorily after irradiation. The signal-to-noise ratio is not seen to change up to fluence of 6 x 10^14 n_eq /cm^2 . The signal time jitter was estimated as the ratio between the voltage noise and the signal slope at threshold. At -35 {^\circ}C, sensor bias voltage of 200 V and frontend power consumption of 0.9 W/cm^2, the time jitter of the most-probable signal amplitude was estimated to be 21 ps for proton fluence up to 6 x 10 n_eq/cm^2 and 57 ps at 1 x 10^16 n_eq/cm^2 . Increasing the sensor bias to 250 V and the analog voltage of the preamplifier from 1.8 to 2.0 V provides a time jitter of 40 ps at 1 x 10^16 n_eq/cm^2., Comment: Submitted to JINST

We present experimental results and a theoretical model that illustrate how competing eigenbases can determine the dynamics of a fluorescing atom. In the absence of a magnetic field, the atom can get trapped in a dark state, which inhibits fluorescence. In general, this will happen when the magnetic degeneracy of the ground state is greater than the one of the excited state. A canonical way to avoid optical pumping to dark states is to apply a magnetic field at an angle with respect to the polarization of the exciting light. This generates a competition of eigenbases which manifests as a crossover between two regimes dominated either by the laser or the magnetic field. We illustrate this crossover with fluorescence measurements on a single laser-cooled calcium ion in a Paul trap and find that it occurs at a critical laser intensity that is proportional to the external magnetic field. We contrast our results with numerical simulations of the atomic levels involved and also present a simple theoretical model that provides excellent agreement with experimental results and facilitates the understanding of the dynamics., Comment: 7 pages, 5 figures

The $R$-dependence of jet observables provides a new tool in understanding the interplay between the jet energy-loss mechanism and medium response in heavy-ion collisions. This work applies the Monte Carlo events generator JEWEL and PYTHIA, coupled with $\rm T_{R}ENTo$ initial conditions and the state-of-the-art (2+1)D v-USPhydro, for the simulation of jet distributions and substructure observables for lead-lead collisions at LHC energy scales. We present the jet nuclear modification $R_{AA}$ and anisotropic flow coefficients $v_{n=2,3}$ varying the jet cone radius $R$, in the context of anti-$k_T$ jets, in addition to leading subjet fragmentation. The calculations indicate the impacts of the hydrodynamic evolution and weakly-coupled medium response, given by recoils, on the distributions. Results are compared to experimental data in a wide range of jet $p_T$ and collision centrality, and displayed along large jets ($R \ge 0.6$) predictions., Comment: 6 pages, 4 figures, contribution to the proceedings of the 11th International Conference on Hard and Electromagnetic Probes of High-Energy Nuclear Collisions (Hard Probes 2023), Aschaffenburg (Germany), March 26-31, 2023. Submitted to PoS

In 5G cellular networks, frequency range 2 (FR2) introduces higher frequencies that cause rapid signal degradation and challenge user mobility. In recent studies, a conditional handover procedure has been adopted as an enhancement to baseline handover to enhance user mobility robustness. In this article, the mobility performance of conditional handover is analyzed for a 5G mm-wave network in FR2 that employs beamforming. In addition, a resource-efficient random access procedure is proposed that increases the probability of contention-free random access during a handover. Moreover, a simple yet effective decision tree-based supervised learning method is proposed to minimize the handover failures that are caused by the beam preparation phase of the random access procedure. Results have shown that a tradeoff exists between contention-free random access and handover failures. It is also seen that the optimum operation point of random access is achievable with the proposed learning algorithm for conditional handover. Moreover, a mobility performance comparison of conditional handover with baseline handover is also carried out. Results have shown that while baseline handover causes fewer handover failures than conditional handover, the total number of mobility failures in the latter is less due to the decoupling of the handover preparation and execution phases., Comment: 8 pages, 6 figures. Accepted for presentation at the European Wireless 2023 Conference (EW2023), to be held in Rome, Italy. arXiv admin note: text overlap with arXiv:1910.11890

We investigate conditions that guarantee local indicability of groups that admit presentations with the homology of a circle, generalizing a result of J. Howie for two-relator presentations. We apply our results to investigate local indicability of LOT groups and some classes of non-cycle-free Adian presentations, extending previous results in that direction by J. Howie and D. Wise., Comment: 15 pages

Accelerated light has been demonstrated with laser light and diffraction. Within the diffracting field it is possible to identify a portion that carries most of the beam energy, which propagates in a curved trajectory as it would have been accelerated by a gravitational field for instance. Here, we analyze the effects of this kind of acceleration over the entanglement between twin beams produced in spontaneous parametric down-conversion. Our results show that acceleration does not affect entanglement significantly, under ideal conditions. The optical scheme introduced can be useful in the understanding of processes in the boundary between gravitation and quantum physics., Comment: 6 pages, 3 figures

In a standard view of the reinforcement learning problem, an agent's goal is to efficiently identify a policy that maximizes long-term reward. However, this perspective is based on a restricted view of learning as finding a solution, rather than treating learning as endless adaptation. In contrast, continual reinforcement learning refers to the setting in which the best agents never stop learning. Despite the importance of continual reinforcement learning, the community lacks a simple definition of the problem that highlights its commitments and makes its primary concepts precise and clear. To this end, this paper is dedicated to carefully defining the continual reinforcement learning problem. We formalize the notion of agents that "never stop learning" through a new mathematical language for analyzing and cataloging agents. Using this new language, we define a continual learning agent as one that can be understood as carrying out an implicit search process indefinitely, and continual reinforcement learning as the setting in which the best agents are all continual learning agents. We provide two motivating examples, illustrating that traditional views of multi-task reinforcement learning and continual supervised learning are special cases of our definition. Collectively, these definitions and perspectives formalize many intuitive concepts at the heart of learning, and open new research pathways surrounding continual learning agents., Comment: NeurIPS 2023

When has an agent converged? Standard models of the reinforcement learning problem give rise to a straightforward definition of convergence: An agent converges when its behavior or performance in each environment state stops changing. However, as we shift the focus of our learning problem from the environment's state to the agent's state, the concept of an agent's convergence becomes significantly less clear. In this paper, we propose two complementary accounts of agent convergence in a framing of the reinforcement learning problem that centers around bounded agents. The first view says that a bounded agent has converged when the minimal number of states needed to describe the agent's future behavior cannot decrease. The second view says that a bounded agent has converged just when the agent's performance only changes if the agent's internal state changes. We establish basic properties of these two definitions, show that they accommodate typical views of convergence in standard settings, and prove several facts about their nature and relationship. We take these perspectives, definitions, and analysis to bring clarity to a central idea of the field.

In light of the recent result published by the Fermilab Muon $(g-2)$ experiment, we investigate a simple model that includes particles of fractional electric charges: a colour-singlet fermion and a scalar with charges $2/3e$ and $1/3e$, respectively. The impact of these particles on the anomalous muon's magnetic moment is examined, particularly the restrictions on their Yukawa couplings with the light leptons. Given that lepton flavor violation processes impose stringent constraints on certain scenarios beyond the Standard Model, we asses the one-loop contribution of the new particles to $(g-2)$ in order to identify regions in the parameter space consistent with the Fermilab results and compatible with the current and projected limits on the branching ratio $Br(\mu \rightarrow e \gamma)$. Taking into account the current lower bound for the masses of fractionary charged particles, which is around 634 GeV, we show that the mass of the scalar particle with fractional charge must exceed 1 TeV. In particular, we present some estimatives for double production of the colour-singlet fermion at the 14 TeV LHC. Finally, we also study the validity of our model in light of the QCD lattice results on the muon $(g-2)$., Comment: 17 pages, 6 figures

A new multivariate integer-valued Generalized AutoRegressive Conditional Heteroscedastic process based on a multivariate Poisson generalized inverse Gaussian distribution is proposed. The estimation of parameters of the proposed multivariate heavy-tailed count time series model via maximum likelihood method is challenging since the likelihood function involves a Bessel function that depends on the multivariate counts and its dimension. As a consequence, numerical instability is often experienced in optimization procedures. To overcome this computational problem, two feasible variants of the Expectation-Maximization (EM) algorithm are proposed for estimating parameters of our model under low and high-dimensional settings. These EM algorithm variants provide computational benefits and help avoid the difficult direct optimization of the likelihood function from the proposed model. Our model and proposed estimation procedures can handle multiple features such as modeling of multivariate counts, heavy-taildness, overdispersion, accommodation of outliers, allowances for both positive and negative autocorrelations, estimation of cross/contemporaneous-correlation, and the efficient estimation of parameters from both statistical and computational points of view. Extensive Monte Carlo simulation studies are presented to assess the performance of the proposed EM algorithms. An application to modeling bivariate count time series data on cannabis possession-related offenses in Australia is discussed., Comment: 32pages, 14figures

We study the impact of data sharing policies on cyber insurance markets. These policies have been proposed to address the scarcity of data about cyber threats, which is essential to manage cyber risks. We propose a Cournot duopoly competition model in which two insurers choose the number of policies they offer (i.e., their production level) and also the resources they invest to ensure the quality of data regarding the cost of claims (i.e., the data quality of their production cost). We find that enacting mandatory data sharing sometimes creates situations in which at most one of the two insurers invests in data quality, whereas both insurers would invest when information sharing is not mandatory. This raises concerns about the merits of making data sharing mandatory., Comment: 46 pages, 8 figures, to be published at Computers & Security

A growing body of evidence suggests that neural networks employed in deep reinforcement learning (RL) gradually lose their plasticity, the ability to learn from new data; however, the analysis and mitigation of this phenomenon is hampered by the complex relationship between plasticity, exploration, and performance in RL. This paper introduces plasticity injection, a minimalistic intervention that increases the network plasticity without changing the number of trainable parameters or biasing the predictions. The applications of this intervention are two-fold: first, as a diagnostic tool $\unicode{x2014}$ if injection increases the performance, we may conclude that an agent's network was losing its plasticity. This tool allows us to identify a subset of Atari environments where the lack of plasticity causes performance plateaus, motivating future studies on understanding and combating plasticity loss. Second, plasticity injection can be used to improve the computational efficiency of RL training if the agent has to re-learn from scratch due to exhausted plasticity or by growing the agent's network dynamically without compromising performance. The results on Atari show that plasticity injection attains stronger performance compared to alternative methods while being computationally efficient., Comment: NeurIPS 2023 camera-ready

In this paper, we implement exponential integrators, specifically Integrating Factor (IF) and Exponential Time Differencing (ETD) methods, using pseudo-spectral techniques to solve phase-field equations within a Python framework. These exponential integrators have showcased robust performance and accuracy when addressing stiff nonlinear partial differential equations. We compare these integrators to the well-known implicit-explicit (IMEX) Euler integrators used in phase-field modeling. The synergy between pseudo-spectral techniques and exponential integrators yields significant benefits for modeling intricate systems governed by phase-field dynamics, such as solidification processes and pattern formation. Our comprehensive Python implementation illustrates the effectiveness of this combined approach in solving phase-field model equations. The results obtained from this implementation highlight the accuracy and computational advantages of the ETD method compared to other numerical techniques., Comment: 12 pages, 7 figures, inline Python code

This tutorial discusses methodology for causal inference using longitudinal modified treatment policies. This method facilitates the mathematical formalization, identification, and estimation of many novel parameters, and mathematically generalizes many commonly used parameters, such as the average treatment effect. Longitudinal modified treatment policies apply to a wide variety of exposures, including binary, multivariate, and continuous, and can accommodate time-varying treatments and confounders, competing risks, loss-to-follow-up, as well as survival, binary, or continuous outcomes. Longitudinal modified treatment policies can be seen as an extension of static and dynamic interventions to involve the natural value of treatment, and, like dynamic interventions, can be used to define alternative estimands with a positivity assumption that is more likely to be satisfied than estimands corresponding to static interventions. This tutorial aims to illustrate several practical uses of the longitudinal modified treatment policy methodology, including describing different estimation strategies and their corresponding advantages and disadvantages. We provide numerous examples of types of research questions which can be answered using longitudinal modified treatment policies. We go into more depth with one of these examples--specifically, estimating the effect of delaying intubation on critically ill COVID-19 patients' mortality. We demonstrate the use of the open-source R package lmtp to estimate the effects, and we provide code on https://github.com/kathoffman/lmtp-tutorial.

We report VLA B-configuration observations of the HI 21 cm line on the close disk galaxy pair NGC 5595 and NGC 5597. At the angular resolution of the observations, $\sim7.1'' \times 4.2''$, while most of the HI 21 cm in NGC 5595 and in NGC 5597 has the same extent as the optical disk, we have detected for the first time extended structures (streamers) to the north-east (NE), and south-west (SW) of NGC 5595 with no counterparts in blue, red optical (continuum), 20 cm radio continuum, or H$\alpha$ spectral-line emission. One structure is extended by $\sim 45''$ to the NE with blue-shifted velocities, and the other by $\sim 20''$ to the SW with red-shifted velocities with respect to the systemic velocity. No HI 21 cm emission is detected from the innermost central (nuclear) regions of either galaxy. Lower angular resolution HI 21 cm imaging indicates the non-existence of any intergalactic HI 21 cm gas as tails or bridges between the two galaxies. Our new 20 cm radio continuum emission image of NGC 5597 shows a strong unresolved elongated structure at the central region, in the north-east south-west direction, very similar to the spatial location of the innermost H$\alpha$ spectral line emission. There is no 20 cm continuum emission from its north spiral arm. In NGC 5595, the 20 cm radio continuum image shows no continuum emission from the NE nor the SW extended structures with HI 21 cm emission., Comment: 22 pages, 14 figures. AJ accepted

Integrated photonics is a powerful contender in the race for a fault-tolerant quantum computer, claiming to be a platform capable of scaling to the necessary number of qubits. This necessitates the use of high-quality quantum states, which we create here using an all-around high-performing photon source on an integrated photonics platform. We use a photonic molecule architecture and broadband directional couplers to protect against fabrication tolerances and ensure reliable operation. As a result, we simultaneously measure a spectral purity of $99.1 \pm 0.1$ %, a pair generation rate of $4.4 \pm 0.1$ MHz mW$^{-2}$, and an intrinsic source heralding efficiency of $94.0 \pm 2.9$ %. We also see a maximum coincidence-to-accidental ratio of $1644 \pm 263$. We claim over an order of magnitude improvement in the trivariate trade-off between source heralding efficiency, purity and brightness. Future implementations of the source could achieve in excess of $99$ % purity and heralding efficiency using state-of-the-art propagation losses., Comment: 8 pages, 4 figures (main text), 11 pages 14 figures (supplementary text), pre-print

This study employs classical Density Functional Theory (cDFT) to investigate the adsorption isotherms and structural information of H2 and CH4 fluids inside MOF-5. The results indicate that the adsorption of both fluids is highly dependent on the fluid temperature and the shape of the MOF-5 structure. Specifically, the CH4 molecules exhibit stronger interactions with the MOF-5 framework, resulting in a greater adsorbed quantity compared to H2. Additionally, the cDFT calculations reveal that the adsorption process is influenced by the fluid-fluid spatial correlations between the fluid molecules and the external potential produced by the MOF-5 solid atoms. These findings are supported by comparison with experimental data of adsorbed amount and the structure factor of the adsorbed fluid inside the MOF-5. We demonstrate the importance of choosing the appropriate grid size in calculating the adsorption isotherm and the fluid structure factors within the MOF-5. Overall, this work provides valuable insights into the adsorption mechanism of H2 and CH4 in MOF-5, emphasizing the importance of considering the structural properties of the adsorbed fluids in MOFs for predicting and designing their gas storage capacity at different thermodynamic conditions., Comment: 13 pages, 11 figures, 2 supplementary movies, Published in Fluid Phase Equilibria Journal

The MALTA CMOS monolithic silicon pixel sensors has been developed in the Tower 180 nm CMOS imaging process. It includes an asynchronous readout scheme and complies with the ATLAS inner tracker requirements for the HL-LHC. Several 4-chip MALTA modules have been built using Al wedge wire bonding to demonstrate the direct transfer of data from chip-to-chip and to read out the data of the entire module via one chip only. Novel technologies such as Anisotropic Conductive Films (ACF) and nanowires have been investigated to build a compact module. A lightweight flex with 17 {\mu}m trace spacing has been designed, allowing compact packaging with a direct attachment of the chip connection pads to the flex using these interconnection technologies. This contribution shows the current state of our work towards a flexible, low material, dense and reliable packaging and modularization of pixel detectors., Comment: 5 pages + 1 page references,8 figures

In this article we fully classify regular tubular surfaces in Euclidean, Lorentzian and hyperbolic 3-spaces whose Gaussian and mean curvatures $K$ and $H$ verify a polynomial relation. More precisely, we determine the set $S(Q)$ of all regular tubular surfaces whose curvatures verify a given polynomial relation $Q(K,H)=0$, and the set $Q(S)$ of all polynomial relations vanished by the curvatures of a given tubular surface $S$.

In this paper we present a method to measure transient fluorescent dynamics with single trapped ions in a Paul trap. We use $^{40}$Ca$^+$ ions which exhibit a $\Lambda$-type three-level system and measure the characteristic optical pumping times between the ground $S_{1/2}$ and the meta-stable $D_{3/2}$ levels. We prepare one of these states, then pump it to the opposite via the excited $P_{1/2}$ state. By measuring the scattered photons of the ion, we retrieve transient curves of the atomic fluorescence. These curves provide fundamental information about the atomic system, such as branching fractions and excited-state lifetimes, as well as experimental parameters like the efficiency of the detection system and the saturation parameter of one of the transitions. Finally, we study the time dependent fluorescence as a function of optical power and detuning of the lasers and find a very good agreement with simulating the dynamics via a three-level open quantum system through a set of optical Bloch equations. Being able to record time dependent fluorescence is of particular interest as it contains information about the temperature, cooling and heating dynamics of the ion., Comment: 9 pages, 3 figures, to be published in JOSA B

In this article, we propose metamaterials analog brane models based on the geometric optics approach. We show how to model the Randall-Sundrum thin brane model and the Gremm model for a thick brane. We incorporate the Yukawa-like confinement mechanics for thick branes in the analog system and show an asymmetrical analog brane model with and without the confinement mechanisms.

A second monolithic silicon pixel prototype was produced for the MONOLITH project. The ASIC contains a matrix of hexagonal pixels with 100 {\mu}m pitch, readout by a low-noise and very fast SiGe HBT frontend electronics. Wafers with 50 {\mu}m thick epilayer of 350 {\Omega}cm resistivity were used to produce a fully depleted sensor. Laboratory and testbeam measurements of the analog channels present in the pixel matrix show that the sensor has a 130 V wide bias-voltage operation plateau at which the efficiency is 99.8%. Although this prototype does not include an internal gain layer, the design optimised for timing of the sensor and the front-end electronics provides a time resolutions of 20 ps., Comment: 11 pages, 11 figures

For each $n\ge 1$ we determine the minimum number of points in a poset with cyclic automorphism group of order $n$., Comment: 12 pages, 5 figures

We present a method of tomography that measures the joint spectral phase (JSP) of spontaneously emitted photon pairs originating from a largely uncharacterized ``target" source. We use quantum interference between our target source and a reference source to extract the JSP with four spectrally resolved measurements, in a process that we call quantum-referenced spontaneous emission tomography (Q-SpET). We have demonstrated this method on a photonic integrated circuit for a target micro-ring resonator photon-pair source. Our results show that spontaneously emitted photon pairs from a micro-ring resonator are distinctively different from that of stimulated emission, and thus cannot in general be fully characterized using classical stimulated emission tomography without detailed knowledge of the source.