Search

Time series forecasting plays a crucial role in diverse fields, necessitating the development of robust models that can effectively handle complex temporal patterns. In this article, we present a novel feature selection method embedded in Long Short-Term Memory networks, leveraging a multi-objective evolutionary algorithm. Our approach optimizes the weights and biases of the LSTM in a partitioned manner, with each objective function of the evolutionary algorithm targeting the root mean square error in a specific data partition. The set of non-dominated forecast models identified by the algorithm is then utilized to construct a meta-model through stacking-based ensemble learning. Furthermore, our proposed method provides an avenue for attribute importance determination, as the frequency of selection for each attribute in the set of non-dominated forecasting models reflects their significance. This attribute importance insight adds an interpretable dimension to the forecasting process. Experimental evaluations on air quality time series data from Italy and southeast Spain demonstrate that our method substantially improves the generalization ability of conventional LSTMs, effectively reducing overfitting. Comparative analyses against state-of-the-art CancelOut and EAR-FS methods highlight the superior performance of our approach.

Quasi-classical trajectory (QCT) calculations were conducted on the newly developed full-dimensional potential energy surface, PES-2023, to analyse two critical aspects: the influence of vibrational versus translational energy in promoting reactivity, and the impact of vibrational excitation within similar vibrational modes. The former relates to Polanyi's rules, while the latter concerns mode selectivity. Initially, the investigation revealed that independent vibrational excitation by a single quantum of ethane's symmetric and asymmetric stretching modes (differing by only 15 cm-1) yielded comparable dynamics, reaction cross-sections, HCN vibrational product distributions, and scattering distributions. This observation dismisses any significant mode selectivity. Moreover, providing an equivalent amount of energy as translational energy (at total energies of 9.6 and 20.0 kcal/mol) gave rise to slightly lower reactivity compared to putting the same amount of energy as vibrational energy. This effect is more evident at low energies, presenting a counterintuitive scenario in an "early transition state" reaction. Regarding CN vibrational excitation, our calculations revealed that the reaction cross-section remains nearly unaffected by this vibrational excitation, suggesting the CN stretching mode is a spectator mode. The results were rationalized by considering several factors: the strong coupling between different vibrational modes, between vibrational modes and the reaction coordinate, and a significant vibrational energy redistribution within the ethane reactant before collision. This redistribution creates an unphysical energy flow, resulting in the loss of adiabaticity and vibrational memory before the reactants' collision.

\textit{Holey Graphene} (HG) is a widely used graphene material for the synthesis of high-purity and highly crystalline materials. In this work, we explore the electronic properties of a periodic distribution of lattice holes, demonstrating the emergence of flat bands with compact localized states. It is shown that the holes break the bipartite sublattice and inversion symmetries, inducing gaps and a nonzero Berry curvature. Moreover, the folding of the Dirac cones from the hexagonal Brillouin zone (BZ) to the holey superlattice rectangular BZ of HG with sizes proportional to an integer $n$ times the graphene's lattice parameter leads to a periodicity in the gap formation such that $n \equiv 0$ (mod $3$). Meanwhile, it is shown that if $n \equiv \pm 1$ (mod $3$), a gap emerges where Dirac points are folded along the $\Gamma-X$ path. The low-energy hamiltonian for the three central bands is also obtained, revealing that the system behaves as an effective $\alpha-\mathcal{T}_{3}$ graphene material. Therefore, a simple protocol is presented here that allows obtaining flat bands at will. Such bands are known to increase electron-electron correlated effects. This work provides an alternative system, much easier to build than twisted systems, to obtain highly correlated quantum phases.

The tunneling potential formalism makes it easy to construct exact solutions to the vacuum decay problem in potentials with multiple fields. While some exact solutions for single-field decays were known, we present the first nontrivial analytic examples with two and three scalar fields, and show how the method can be generalized to include gravitational corrections. Our results illuminate some analytic properties of the tunneling potential functions and can have a number of uses, among others: to serve as simple approximations to realistic potentials; to learn about parametric dependencies of decay rates; to check conjectures on vacuum decay; as benchmarks for multi-field numerical codes; or to study holographic interpretations of vacuum decay., Comment: 17 pages

Temperature dependence of the thermal rate constants and kinetic isotope effects (KIE) of the CN + C2H6 gas-phase hydrogen abstraction reaction was theoretically determined within the 25-1000 K temperature range, i.e., from ultra-low to high-temperature regimes. Based on a recently developed full-dimensional analytical potential energy surface fitted to highly accurate explicitly correlated ab initio calculations, three different kinetic theories were used: canonical variational transition state theory (CVT), quasiclassical trajectory theory (QCT), and ring polymer molecular dynamics (RPMD) method for the computation of rate constants. We found that the thermal rate constants obtained with the three theories show a V-shaped temperature dependence, with a pronounced minimum near 200 K, qualitatively reproducing the experimental measurements. Among the three methods used in this work, the QCT and RPMD methods have the best agreement with the experiment at low and high temperatures, respectively. The significant increase in the rate constant at ultra-low temperatures in this very exothermic and practically barrierless reaction can be attributed to the large value of the impact parameter, ruling out the role of the tunneling effect and the intermediate complexes in the entrance channel. The theoretical H/D KIE depicted a normal behaviour, i.e., values greater than unity, emulating the experimental measurements and previous theoretical results. Finally, the discrepancies between theory and experiments were analysed as a function of several factors, such as limitations of the kinetics theories and the potential energy surface, as well as the uncertainties in the experimental measurements.

The Bestest Little Higgs Model (BLHM) has attracted increasing attention in recent years, mainly because it can explain the hierarchy problem without fine-tuning by introducing one-loop corrections to the Higgs boson mass through heavy top-quark partners and gauge bosons. In this paper, we lay the groundwork for future research in the BLHM, derive and present Feynman rules for scalars ($h_0, H_0, \phi^{0}, \eta^{0}, \sigma, H^{\pm}, \phi^{\pm}, \eta^{\pm}, A_0$) and include the self-interactions of Higgs bosons. We also provide the Mathematica notebooks used to obtain the Feynman rules. These Feynman rules help investigate new physics beyond the Standard Model. The calculations are performed using FeynCalc, and all model files are publicly available online at http://ricaxcan.uaz.edu.mx/jspui/simple-search. Export of the Lagrangians to FeynArt is also posible., Comment: 32 pages, 0 figure

We generalize results from topological robotics on the topological complexity (TC) of aspherical spaces to sectional categories of fibrations inducing subgroup inclusions on the level of fundamental groups. In doing so, we establish new lower bounds on sequential TCs of aspherical spaces as well as the parametrized TC of epimorphisms. Moreover, we generalize the Costa-Farber canonical class for TC to classes for sequential TCs and explore their properties. We combine them with the results on sequential TCs of aspherical spaces to obtain results on spaces that are not necessarily aspherical., Comment: 40 pages

Bubbles of nothing (BoNs) describe the decay of spacetimes with compact dimensions and are thus of fundamental importance for many higher dimensional theories proposed beyond the Standard Model. BoNs admit a 4-dimensional description in terms of a singular Coleman-de Luccia (CdL) instanton involving the size modulus field, stabilized by some potential $V(\phi)$. Using the so-called tunneling potential ($V_t$) approach, we study which types of BoNs are possible and for which potentials $V(\phi)$ can they be present. We identify four different types of BoN, characterized by different asymptotic behaviours at the BoN core and corresponding to different classes of higher dimensional theories, which we also classify. Combining numerous analytical and numerical examples, we study the interplay of BoN decays with other standard decay channels, identify the possible types of quenching of BoN decays and show how BoNs for flux compactifications can also be described in 4 dimensions by a multifield $V_t$. The use of the $V_t$ approach greatly aids our analyses and offers a very simple picture of BoNs which are treated in the same language as any other standard vacuum decays., Comment: 53 pages + appendices

Machine learning (ML) is emerging as a transformative tool for the design of architected materials, offering properties that far surpass those achievable through lab-based trial-and-error methods. However, a major challenge in current inverse design strategies is their reliance on extensive computational and/or experimental datasets, which becomes particularly problematic for designing micro-scale stochastic architected materials that exhibit nonlinear mechanical behaviors. Here, we introduce a new end-to-end scientific ML framework, leveraging deep neural operators (DeepONet), to directly learn the relationship between the complete microstructure and mechanical response of architected metamaterials from sparse but high-quality in situ experimental data. The approach facilitates the inverse design of structures tailored to specific nonlinear mechanical behaviors. Results obtained from spinodal microstructures, printed using two-photon lithography, reveal that the prediction error for mechanical responses is within a range of 5 - 10%. Our work underscores that by employing neural operators with advanced micro-mechanics experimental techniques, the design of complex micro-architected materials with desired properties becomes feasible, even in scenarios constrained by data scarcity. Our work marks a significant advancement in the field of materials-by-design, potentially heralding a new era in the discovery and development of next-generation metamaterials with unparalleled mechanical characteristics derived directly from experimental insights., Comment: 29 pages, 5 figures

In this paper we study the joint determination of source and background flux for point sources as observed by digital array detectors. We explicitly compute the two-dimensional Cram\'er-Rao absolute lower bound (CRLB) as well as the performance bounds for high-dimensional implicit estimators from a generalized Taylor expansion. This later approach allows us to obtain computable prescriptions for the bias and variance of the joint estimators. We compare these prescriptions with empirical results from numerical simulations in the case of the weighted least squares estimator (introducing an improved version, denoted stochastic weighted least-squares) as well as with the maximum likelihood estimator, finding excellent agreement. We demonstrate that these estimators provide quasi-unbiased joint estimations of the flux and background, with a variance that approaches the CRLB very tightly and are, hence, optimal, unlike the case of sequential estimation used commonly in astronomical photometry which is sub-optimal. We compare our predictions with numerical simulations of realistic observations, as well as with observations of a bona-fide non-variable stellar source observed with TESS, and compare it to the results from the sequential estimation of background and flux, confirming our theoretical expectations. Our practical estimators can be used as benchmarks for general photometric pipelines, or for applications that require maximum precision and accuracy in absolute photometry., Comment: Accepted by PASP. 35 pages, 12 figures, 5 appendices

We argue for the need for a new generation of data science solutions that can democratize recent advances in data engineering and artificial intelligence for non-technical users from various disciplines, enabling them to unlock the full potential of these solutions. To do so, we adopt an approach whereby computational creativity and conversational computing are combined to guide non-specialists intuitively to explore and extract knowledge from data collections. The paper introduces MATILDA, a creativity-based data science design platform, showing how it can support the design process of data science pipelines guided by human and computational creativity.

Mechanical metamaterials, whose unique mechanical properties stem from their structural design rather than material constituents, are gaining popularity in engineering applications. In particular, recent advances in self-assembly techniques offer the potential to fabricate load-bearing mechanical metamaterials with unparalleled feature size control and scalability compared to those produced by additive manufacturing (AM). Yet, the field is still in its early stages. In this perspective, we first provide an overview of the state-of-the-art self-assembly techniques, with a focus on the copolymer and colloid crystal self-assembly processes. We then discuss current challenges and future opportunities in this research area, focusing on novel fabrication approaches, the need for high-throughput characterization methods, and the integration of Machine Learning (ML) and lab automation for inverse design. Given recent progress in all these areas, we foresee mechanical metamaterials fabricated from self-assembly techniques impacting a variety of applications relying on lightweight, strong, and tough materials., Comment: 24 pages, 3 figures

This paper proposes a conversational approach implemented by the system Chatin for driving an intuitive data exploration experience. Our work aims to unlock the full potential of data analytics and artificial intelligence with a new generation of data science solutions. Chatin is a cutting-edge tool that democratises access to AI-driven solutions, empowering non-technical users from various disciplines to explore data and extract knowledge from it.

Stress-strain curves, or more generally, stress functions, are an extremely important characterization of a material's mechanical properties. However, stress functions are often difficult to derive and are narrowly tailored to a specific material. Further, large deformations, high strain-rates, temperature sensitivity, and effect of material parameters compound modeling challenges. We propose a generalized deep neural network approach to model stress as a state function with quantile regression to capture uncertainty. We extend these models to uniaxial impact mechanics using stochastic differential equations to demonstrate a use case and provide a framework for implementing this uncertainty-aware stress function. We provide experiments benchmarking our approach against leading constitutive, machine learning, and transfer learning approaches to stress and impact mechanics modeling on publicly available and newly presented data sets. We also provide a framework to optimize material parameters given multiple competing impact scenarios., Comment: Index Terms: Stress, Uncertainty, Impact Mechanics, Deep Learning, Neural Network. 10 pages, 9 figures, 6 tables

Metaphor identification aims at understanding whether a given expression is used figuratively in context. However, in this paper we show how existing metaphor identification datasets can be gamed by fully ignoring the potential metaphorical expression or the context in which it occurs. We test this hypothesis in a variety of datasets and settings, and show that metaphor identification systems based on language models without complete information can be competitive with those using the full context. This is due to the construction procedures to build such datasets, which introduce unwanted biases for positive and negative classes. Finally, we test the same hypothesis on datasets that are carefully sampled from natural corpora and where this bias is not present, making these datasets more challenging and reliable., Comment: Short paper accepted to EMNLP 2023 main conference

Concepts play a central role in many applications. This includes settings where concepts have to be modelled in the absence of sentence context. Previous work has therefore focused on distilling decontextualised concept embeddings from language models. But concepts can be modelled from different perspectives, whereas concept embeddings typically mostly capture taxonomic structure. To address this issue, we propose a strategy for identifying what different concepts, from a potentially large concept vocabulary, have in common with others. We then represent concepts in terms of the properties they share with the other concepts. To demonstrate the practical usefulness of this way of modelling concepts, we consider the task of ultra-fine entity typing, which is a challenging multi-label classification problem. We show that by augmenting the label set with shared properties, we can improve the performance of the state-of-the-art models for this task., Comment: Accepted for EMNLP 2023

Despite its relevance, the maturity of NLP for social media pales in comparison with general-purpose models, metrics and benchmarks. This fragmented landscape makes it hard for the community to know, for instance, given a task, which is the best performing model and how it compares with others. To alleviate this issue, we introduce a unified benchmark for NLP evaluation in social media, SuperTweetEval, which includes a heterogeneous set of tasks and datasets combined, adapted and constructed from scratch. We benchmarked the performance of a wide range of models on SuperTweetEval and our results suggest that, despite the recent advances in language modelling, social media remains challenging., Comment: EMNLP 2023 Findings

A study of strongly curved graphene magnetization and magnetic susceptibility is carried out. Through a Dirac model complemented with a tight-binding model analysis, we are able to show that mechanical deformations solve the long-standing problem of graphene's theoretically calculated diamagnetic divergence at low temperatures. This suggests that corrugations and mechanical defects in graphene are the cause of finite experimentally measurable magnetic susceptibility. Furthermore, a mechanical effect is also found due to an electronic contribution, which produces a pseudo-de Haas van Alphen (dHvA) effect. This effect is related to oscillating (electronic) forces that oppose deformations; these forces are divergent in flat graphene, indicating that graphene (without substrate) achieves mechanical equilibrium by corrugations. In addition, paramagnetism is predicted for graphene with negative curvature under strong magnetic fields., Comment: 19 pages, 7 figures

This paper presents the results of 475 hours of interferometric observations with the Australia Telescope Compact Array towards the Spiderweb protocluster at \(z=2.16\). We search for large, extended molecular gas reservoirs among 46 previously detected CO(1-0) emitters, employing a customised method we developed. Based on the CO emission images and position-velocity diagrams, as well as the ranking of sources using a binary weighting of six different criteria, we have identified 14 robust and 7 tentative candidates that exhibit large extended molecular gas reservoirs. These extended reservoirs are defined as having sizes greater than 40 kpc or super-galactic scale. This result suggests a high frequency of extended gas reservoirs, comprising at least \(30 \%\) of our CO-selected sample. An environmental study of the candidates is carried out based on N-th nearest neighbour and we find that the large molecular gas reservoirs tend to exist in denser regions. The spatial distribution of our candidates is mainly centred on the core region of the Spiderweb protocluster. The performance and adaptability of our method are discussed. We found 13 (potentially) extended gas reservoirs located in nine galaxy (proto)clusters from the literature. We noticed that large extended molecular gas reservoirs surrounding (normal) star-forming galaxies in protoclusters are rare. This may be attributable to the lack of observations low-J CO transitions and the lack of quantitative analyses of molecular gas morphologies. The large gas reservoirs in the Spiderweb protocluster are a potential source of the intracluster medium seen in low redshift Virgo- or Coma-like galaxy clusters., Comment: 25 pages, 16 figures. Accepted for publication in MNRAS

The flavor-violating decay of a new neutral massive gauge boson $Z^\prime\to\mu e$ is analyzed in the context of extended models, in which this particle emerges. By means of the analysis of the $\mu\to e\gamma$ decay, $\mu-e$ conversion process in nuclei and the $\mu \to e e^{+}e^{-}$ decay, the strength of the $Z^\prime\mu e$ coupling is estimated and used to calculate the branching ratio of the $Z^\prime\to \mu e$ decay. This is done for the so-called $Z_S,\,Z_{LR},\,Z_\chi, Z_\psi$ and $Z_\eta$ bosons. We found that, through the $\mu-e$ conversion process, the most restrictive bound for the coupling is provided by the $Z_{LR}$ boson. Meanwhile, by means of the $\mu \to e e^{+}e^{-}$ decay, the most restrictive bound for the $Z^\prime\mu e$ coupling is provided by the $Z_\chi$ boson. However, if we concentrate on the less restrictive prediction for the Br($Z^\prime\to \mu e$), this comes from the $Z_\eta$ boson and the resulting branching ratio is less than $10^{-4}$. On the other hand, if we consider the most restrictive bound, the branching ratio for the process is below $2\times 10^{-7}$, which results from the $Z_{RL}$ boson, and is obtained through the $\mu-e$ conversion process., Comment: 26 pages, 10 figures

An important characteristic of cosmic hydrogen reionization is the growth of ionized gas bubbles surrounding early luminous objects. Ionized bubble sizes are beginning to be probed using Lyman-$\alpha$ emission from high-redshift galaxies, and will also be probed by upcoming 21-cm maps. We present results from a study of bubble sizes using the state-of-the-art THESAN radiation-hydrodynamics simulation suite, which self-consistently models radiation transport and realistic galaxy formation. We employ the mean-free path method, and track the evolution of the effective ionized bubble size at each point ($R_{\rm eff}$) throughout the Epoch of Reionization. We show there is a slow growth period for regions ionized early, but a rapid "flash ionization" process for regions ionized later as they immediately enter a large, pre-existing bubble. We also find that bright sources are preferentially in larger bubbles, and find consistency with recent observational constraints at $z \gtrsim 9$, but tension with idealized Lyman-$\alpha$ damping-wing models at $z \approx 7$. We find that high overdensity regions have larger characteristic bubble sizes, but the correlation decreases as reionization progresses, likely due to runaway formation of large percolated bubbles. Finally, we compare the redshift at which a region transitions from neutral to ionized ($z_{\rm reion}$) with the time it takes to reach a given bubble size and conclude that $z_{\rm reion}$ is a reasonable local probe of small-scale bubble size statistics ($R_\text{eff} \lesssim 1\,\rm{cMpc}$). However, for larger bubbles, the correspondence between $z_{\rm reion}$ and size statistics weakens due to the time delay between the onset of reionization and the expansion of large bubbles, particularly at high redshifts., Comment: 15 pages, 15 figures. Comments welcome. Please visit https://www.thesan-project.com for more details. Accepted for publication in MNRAS

We consider a model of directed polymers on regular trees with complex-valued random weights introduced by Cook and Derrida [CD90] and studied mathematically by Derrida, Evans and Speer [DES93]. In addition to the usual weak-disorder and strong-disorder regimes, the phase diagram of the model contains a third region due to the effects of the random phases that cannot be observed in the model with positive weights. In this work, we extend the results of [DES93] in two directions. First, we remove the hypothesis on the independence of the random radii and random phases of the environment from most of the phase diagram, and second, under mild assumptions on the law of the environment, we strengthen the convergence of the free energy (shown to hold in probability in [DES93]) to an almost sure convergence. Along the way, many of the arguments of [DES93] are simplified.