Consistent Embedding Research Articles

Localised gravity and resolved braneworlds

Deriving an effective massless field theory for fluctuations about a braneworld spacetime requires analysis of the transverse-space-wavefunction’s second-order differential equation. There can be two strikingly different types of effective theory. For a supersymmetric braneworld, one involves a technically consistent embedding of a supergravity theory on the worldvolume; the other can produce, in certain situations, a genuine localisation of gravity near the worldvolume but not via a technically consistent embedding. So, in the latter situation, the theory’s dynamics remains higher-dimensional but there can still be a lower-dimensional effective-theory interpretation of the dynamics at low worldvolume momenta / large worldvolume distances.This paper examines the conditions for such a gravity localisation to be possible. Localising gravity about braneworld spacetimes requires finding solutions to transverse-space self-adjoint Sturm-Liouville problems admitting a normalisable zero mode in the noncompact transverse space. This in turn requires analysis of Sturm-Liouville problems with radial singular endpoints following a formalism originating in the work of Hermann Weyl. Examples of such gravity-localising braneworld systems are found and analysed in this formalism with underlying “skeleton” braneworlds of Salam-Sezgin, resolved D3-brane and Randall-Sundrum II types.

Multi-view attributed graph clustering based on graph diffusion convolution with adaptive fusion

Graph clustering is one of the popular techniques in data mining, which is widely used in social network analysis, recommendation system and anomaly detection. The existing methods based on multi-view attributed graph ignore the difference of different views and the consistency between views. To this end, this paper proposes a novel method for multi-view attributed graph clustering based on graph diffusion convolution with adaptive fusion. Specifically, a graph diffusion convolution is firstly employed to propagate features of the adjacency and attribute matrices on each view. A linear encoder–decoder is then used to learn the graph embedding representation of each view. Meanwhile, the consistency learning module is applied to capture the geometric relationship consistency of different views for finding a consistent embedding space during the embedding representation learning. To measure the importance of different views, the attention mechanism is subsequently utilized to adaptively learn the weights of different views to fuse multiple views of data for the final clustering. Finally, the graph representation learning and clustering tasks are jointly optimized to obtain a clustering friendly graph representation. Experiments conducted on two public datasets and three real futures datasets from the Zhengzhou Commodity Exchange demonstrate the effectiveness of the proposed method in the task of graph clustering.

Weighing Milky Way and Andromeda in an expanding ΛCDM Universe

The dynamics of the Local Group (LG), especially the contribution of the Milky Way (MW) and Andromeda (M 31) galaxies, is sensitive to the presence of dark energy. This work analyzes the evolution of the LG by considering it as a two-body problem in a homogeneous and isotropic expanding spacetime in a full Λcold dark matter (ΛCDM) background. Using the timing argument (TA), which links LG dynamics to LG mass, we find that the full ΛCDM background predicts a ∼10% lower mass for the LG; whereas Λ alone predicts a ∼10% higher mass. The TA mass is modified by (i) simulations and (ii) the effect of the Large Magellanic Cloud (LMC) to alleviate the poorly constrained internal mass distributions of M 31 and the MW, their time evolution, and the unknown distribution of dark matter between them. First, using IllustrisTNG simulations, we accounted for the effects of two extended halos and their environment (rather than point particles) and predicted their mass (3.89 ± 0.62)×1012 M⊙. Second, the LMC effectively changes the separation and velocities of M 31 towards the MW and reduces the predicted mass to (2.33 ± 0.72)×1012 M⊙. Despite the uncertainties around dark matter between these galaxies, the overall estimated mass is compatible with the mere sum of the MW and M 31 masses. The total mass of the TA is compatible with other estimates, such as the Hubble flow and the Virial Theorem with other dwarf galaxies. The combined result shows, for the first time, that a lower mass estimate can be obtained from the TA, with a consistent embedding and other systematic effects, and without an additional dark matter halo around the galaxies.

Incremental minimum distortion embedding for online structural damage classification.

The damage detection problem in structures diminishes the maintenance cost and extends the life of the structures. Online structural damage classification is a relevant topic in structural health monitoring since monitoring the structure using sensors can help decision-makers can rely on empirical evidence for making informed choices regarding the maintenance, repair, or replacement of structural components. Sensors generate high-dimensional data. To address this challenge, employing dimensionality reduction methods becomes essential. These methods help obtain a low-dimensional representation of the data. Traditional dimensionality reduction methods are trained on a fixed data set, and incorporating new data may require retraining the entire model. This study employed the Minimum Distortion Embedding (MDE) method incrementally, ensuring the maintenance of a consistent embedding as new data becomes available. This incremental embedding took advantage of an anchor constraint to pin the existing embedding in place, then embed the new points. An online structural classification methodology was developed using the incremental MDE and a stream based hoeffding tree classifier. Stream data processing differs from batch processing, as it involves the real-time consumption of each incoming data. The developed methodology was tested with data obtained by accelerometers in a laboratory scaled jacket-type wind turbine foundation. The structural damage classification problems had 5 different classes including the undamaged class and 4 different damage classes. The damage corresponded to a 5mm crack in four different structural elements of the wind turbine foundation. The results indicate the good behavior of the damage classification methodology supported by the high values of the classification metrics obtained.

Perceptual dimensions of wood materials.

Materials exhibit an extraordinary range of visual appearances. Characterizing and quantifying appearance is important not only for basic research on perceptual mechanisms but also for computer graphics and a wide range of industrial applications. Although methods exist for capturing and representing the optical properties of materials and how they vary across surfaces (Haindl & Filip, 2013), the representations are typically very high-dimensional, and how these representations relate to subjective perceptual impressions of material appearance remains poorly understood. Here, we used a data-driven approach to characterizing the perceived appearance characteristics of 30 samples of wood veneer using a "visual fingerprint" that describes each sample as a multidimensional feature vector, with each dimension capturing a different aspect of the appearance. Fifty-six crowd-sourced participants viewed triplets of movies depicting different wood samples as the sample rotated. Their task was to report which of the two match samples was subjectively most similar to the test sample. In another online experiment, 45 participants rated 10 wood-related appearance characteristics for each of the samples. The results reveal a consistent embedding of the samples across both experiments and a set of nine perceptual dimensions capturing aspects including the roughness, directionality, and spatial scale of the surface patterns. We also showed that a weighted linear combination of 11 image statistics, inspired by the rating characteristics, predicts perceptual dimensions well.

Gender sensitisation through course design thinking

Translators’ training providers, including Higher Education (HE), play a big part in shaping the minds and behaviours of future translators. Therefore, translation courses must focus on implementing practices and resources which will enhance not only students’ linguistic and technical skills, but also cultural awareness, social responsibility, and ethical principles. Sexual/identity discourse is an area of critical enquiry which presents significant challenges that translators frequently do not seem to be able to handle due to their generally low level of gender understanding and sensitivity. This is where translators’ trainers need to take action. The current article – which collects the first findings of an in-progress educational project carried out at London Metropolitan University – suggests examples of activities and aspects to consider in curriculum design to create an effective framework which can ensure a more consistent embedding of gendersensitive policies in the curriculum of translation courses.

Consistent graph embedding network with optimal transport for incomplete multi-view clustering

Existing incomplete multi-view learning models focus on reconstructing the latent variables of multiple views by exploring complementary and consistent information among diverse views. However, filling the missing information for views results in a loss of consistency, while fusion and reconstruction between views face over-fitting problems. The optimal transport algorithm delicately measures the distance of two distributions, resulting in decreased reconstruction errors and guaranteeing consistency and complementarity across multiple views of the data. In light of this, this paper proposes the incorporation of the optimal transport algorithm into the framework of incomplete multi-view clustering. The proposed consistent graph embedding network (CGEN-OT) via optimal transport combines the adversarial module and fusion module to learn a completed latent graph embedding. Specifically, CGEN-OT utilizes an adversarial module to generate complete views and fuses them into a consistent embedding, and introduces reconstruction loss and Sinkhorn loss to jointly optimize the proposed network and obtain superior latent graph embedding and clustering performance. To further validate the clustering accuracy and convergence of the CGEN-OT, experimental evaluation was conducted on six distinct incomplete datasets. A comparison with existing state-of-the-art models highlights the superiority of the proposed framework.

Semantic Consistent Embedding for Domain Adaptive Zero-Shot Learning.

Unsupervised domain adaptation has limitations when encountering label discrepancy between the source and target domains. While open-set domain adaptation approaches can address situations when the target domain has additional categories, these methods can only detect them but not further classify them. In this paper, we focus on a more challenging setting dubbed Domain Adaptive Zero-Shot Learning (DAZSL), which uses semantic embeddings of class tags as the bridge between seen and unseen classes to learn the classifier for recognizing all categories in the target domain when only the supervision of seen categories in the source domain is available. The main challenge of DAZSL is to perform knowledge transfer across categories and domain styles simultaneously. To this end, we propose a novel end-to-end learning mechanism dubbed Three-way Semantic Consistent Embedding (TSCE) to embed the source domain, target domain, and semantic space into a shared space. Specifically, TSCE learns domain-irrelevant categorical prototypes from the semantic embedding of class tags and uses them as the pivots of the shared space. The source domain features are aligned with the prototypes via their supervised information. On the other hand, the mutual information maximization mechanism is introduced to push the target domain features and prototypes towards each other. By this way, our approach can align domain differences between source and target images, as well as promote knowledge transfer towards unseen classes. Moreover, as there is no supervision in the target domain, the shared space may suffer from the catastrophic forgetting problem. Hence, we further propose a ranking-based embedding alignment mechanism to maintain the consistency between the semantic space and the shared space. Experimental results on both I2AwA and I2WebV clearly validate the effectiveness of our method.

De Sitter space as a BRST invariant coherent state of gravitons

The $S$-matrix formulation indicates that a consistent embedding of the de Sitter state in quantum gravity is possible exclusively as an excited quantum state constructed on top of a valid $S$-matrix vacuum such as the Minkowski vacuum. In the present paper we offer such a construction of the de Sitter state in the form of a coherent state of gravitons. Unlike previous realizations of this idea, we focus on BRST invariance as the guiding principle for physicality. In order to establish the universal rules of gauge consistency, we study the BRST-invariant construction of coherent states created by classical and quantum sources in QED and in linearized gravity. Introduction of $N$ copies of scalar matter coupled to gravity allows us to take a special double scaling limit, a so-called species limit, in which our construction of the de Sitter state becomes exact. In this limit, the irrelevant quantum gravitational effects vanish, whereas the collective phenomena, such as Gibbons-Hawking radiation, are calculable.

\u03b1\u2032-corrections to near extremal dyonic strings and weak gravity conjecture

We construct non-extremal dyonic string solutions in 6D minimal supergravity where the leading higher derivative corrections arise from either the type IIA string theory compactified on K3 or the heterotic string theory compactified on 4-torus. The thermodynamical quantities and Euclidean actions of the strings are computed. In the near extremal regime, we calculate the force felt by a probe fundamental string in the background of the macroscopic dyonic string with leading α′ corrections. We find that in both the IIA and heterotic setups, away from extremality, the attractive force overwhelms the repulsive force. However, close to extremality, the α′ corrections can reduce the attractive force in the isoentropic process, where the charges are fixed. This feature may be used as a new constraint for supergravity models with consistent quantum gravity embedding, in cases where the extremal limit coincides with the BPS limit and the higher derivative corrections do not affect the mass-to-charge ratio. By contrast, the α′ corrections can enhance the attractive force in the isothermal or isoenergetic processes.

Analytic and exponentially localized braneworld Reissner-Nordström-AdS solution: A top-down approach

In this work, we construct a five-dimensional spherically-symmetric, charged and asymptotically Anti-de Sitter black hole with its singularity being point-like and strictly localised on our brane. In addition, the induced brane geometry is described by a Reissner-Nordstr\"{o}m-(A)dS line-element. We perform a careful classification of the horizons, and demonstrate that all of them are exponentially localised close to the brane thus exhibiting a pancake shape. The bulk gravitational background is everywhere regular, and reduces to an AdS$_5$ spacetime right outside the black-hole event horizon. This geometry is supported by an anisotropic fluid with only two independent components, the energy density $\rho_E$ and tangential pressure $p_2$. All energy conditions are respected close to and on our brane, but a local violation takes place within the event horizon regime in the bulk. A tensor-vector-scalar field-theory model is built in an attempt to realise the necessary bulk matter, however, in order to do so, both gauge and scalar degrees of freedom need to turn phantom-like at the bulk boundary. The study of the junction conditions reveals that no additional matter needs to be introduced on the brane for its consistent embedding in the bulk geometry apart from its constant, positive tension. We finally compute the effective gravitational equations on the brane, and demonstrate that the Reissner-Nordstr\"{o}m-(A)dS geometry on our brane is caused by the combined effect of the five-dimensional geometry and bulk matter with its charge being in fact a tidal charge.

Mapping functional gradients of the striatal circuit using simultaneous microelectric stimulation and ultrahigh-field fMRI in non-human primates

Advances in functional magnetic resonance imaging (fMRI) have significantly enhanced our understanding of the striatal system of both humans and non-human primates (NHP) over the last few decades. However, its circuit-level functional anatomy remains poorly understood, partly because in-vivo fMRI cannot directly perturb a brain system and map its casual input-output relationship. Also, routine 3T fMRI has an insufficient spatial resolution. We performed electrical microstimulation (EM) of the striatum in lightly-anesthetized NHPs while simultaneously mapping whole-brain activation, using contrast-enhanced fMRI at ultra-high-field 7T. By stimulating multiple positions along the striatum's main (dorsal-to-ventral) axis, we revealed its complex functional circuit concerning mutually connected subsystems in both cortical and subcortical areas. Indeed, within the striatum, there were distinct brain activation patterns across different stimulation sites. Specifically, dorsal stimulation revealed a medial-to-lateral elongated shape of activation in upper caudate and putamen areas, whereas ventral stimulation evoked areas confined to the medial and lower caudate. Such dorsoventral gradients also appeared in neocortical and thalamic activations, indicating consistent embedding profiles of the striatal system across the whole brain. These findings reflect different forms of within-circuit and inter-regional neuronal connectivity between the dorsal and ventromedial striatum. These patterns both shared and contrasted with previous anatomical tract-tracing and in-vivo resting-state fMRI studies. Our approach of combining microstimulation and whole-brain fMRI mapping in NHPs provides a unique opportunity to integrate our understanding of a targeted brain area's meso- and macro-scale functional systems.

Localized brane-world black hole analytically connected to an AdS5 boundary

We construct from first principles the geometry of an analytic, exponentially localized five-dimensional brane-world black hole. The black-hole singularity lies entirely on the 3-brane, while the event horizon is shown to have a pancake shape. The induced line-element on the brane assumes the form of the Schwarzschild solution while the bulk geometry is effectively AdS5 outside the horizon. The derived geometry is supported by an anisotropic fluid in the bulk described only by two independent components, the energy density and tangential pressure, whereas no matter needs to be introduced on the brane for its consistent embedding in the bulk.

Quantum line operators from Lax pairs

Motivated by the realization of the Yang–Baxter equation of 2D integrable models in the 4D gauge theory of Costello–Witten–Yamazaki (CWY), we study the embedding of integrable 2D Toda field models inside this construction. This is done by using the Lax formulation of 2D integrable systems and by thinking of the standard Lax pair L± in terms of components of CWY gauge connection, propagating along particular directions in the gauge bundle. We also use the results of the CWY theory to build quantum line operators for 2D Toda models and compute the one loop contribution of two intersecting lines exchanging one gluon. Other features such as local symmetries and comments on extension of our method to other 2D integrable models are also discussed. We also comment on some basic points that still need a refinement before discussing about a fully consistent embedding of Lax pairs into CWY theory.

Copyright Protection for Holographic Video Using Spatiotemporal Consistent Embedding Strategy

The interest in a three-dimensional (3-D) holographic display is increasing daily because it can generate full depth cue without utilizing any special glasses. However, the content of 3-D holographic videos currently without any protection can be illegally distributed and maliciously manipulated. Therefore, there is an urgent need to protect the copyright of holographic videos against malicious copying. Since watermarking is an effective way to protect the ownership of holographic sequence, this paper proposes a spatiotemporal consistent embedding algorithm for the holographic video watermarking. Imperceptibility requirement in the holographic video watermarking is more challenging compared with static holograms because of the temporal dimension existing in videos. The embedding algorithm should not only consider spatially embedding strength for each frame of the video, but also take the temporal dimension into account in order to guarantee the visual quality of the moving object. Before embedding, to defend the imperceptibility of the watermark from the holographic moving object, the embedding parameters are evaluated by the salient object from the interframe and intraframe. Different from the previous video watermarking algorithm, in order to ensure the robustness, in our paper, instead of two-dimensional watermark, 3-D watermark converted data (QR code) are embedded in the cellular automata (CA) domains using 3-D CA filters. Finally, the QR codes can be extracted from the watermarked holographic frames, and the final 3-D watermark can be digitally reconstructed with different depths’ cue using the computational integral imaging reconstruction algorithm. The experimental results demonstrate that the proposed method exhibits superior performance compared to several methods in the literature, especially the robustness against additive noise and compression attacks.

CLIL PRACTICES IN TOMSK POLYTECHNIC UNIVERSITY: SUCCESSES AND FAILURES

Introduction. Content and Language Integrated Learning (CLIL) has recently been applied into education system, but a range of supporters is rapidly increasing, covering diverse countries. The CLIL approaches simultaneously use a foreign language as an object to studying and as an instrument to learning other subjects. The emergence of CLIL was related to a wide distribution of bilingualism in society and the need to cope with the problems related to this process, including the elimination of international conflicts, the consolidation of society as the factor of a social well-being, the support of a competitive labour market, etc.Despite growing popularity of CLIL across the world and recognised status by the European Union as one of the leading effective means to implementing international language policy of multilingualism, the academic teaching staff of Russian higher school does not still fully understand the organisational mechanisms and forms of a similar learning process, largely as a result of national characteristics and traditions. In a globalising world, CLIL didactics is currently becoming a perspective direction of higher education and an effective way both of acquiring and improving language and communicative competencies to provide further successful professionalising of students and to facilitate their career promotion in the chosen field of activity.Theaimof this paper was to sum up and discuss the experience of Tomsk Polytechnic University in conducting CLIL-based courses.Methodology and research methods. The concept of practice-oriented vocational education was used as the methodological base of the experimental work in order to search optimal variants when developing foreign language knowledge in the students of engineering specialties. In the course of the research, the approaches of comparative analysis, synthesis, generalisation, forecasting, design, modelling, prolonged monitoring of learning outcomes, observation and questionnaire survey were applied.Results and scientific novelty.The actualising reasons for CLIL approach in the Russian high school education were formulated and proved. The brief historical overview of didactic concepts was presented. The authors analysed temporary challenges which preceded the implementation of CLIL methodologies in the learning process of Tomsk Polytechnic University (TPU) and the introduction of programme “Vocational Training and Education in a Foreign Language” (3rd-4th years of education). The stages of teaching foreign language evolution at the university were distinguished: the anglification of engineering education transformed into multidisciplinary model of English for Specific Purpose (ESP); the creation of “pedagogical tandems” – the organization of courses with “double agents”; and, finally, CLIL training in which a cognitive component can be seen as the key characteristic and the main advantage. Success and failure in the field of CLIL practices available today at TPU were revised. The examples of private implementable practices were described and analysed. The research findings revealed the rationality when using various applied variations of CLIL methodology.Practical significance. The research outcomes allowed the authors to adjust not only the main educational programmes of higher education institution, but also internal programmes of professional development of teachers. Although the long-term experiment summarised in the publication in a single higher education institution does not apply for universality and mass circulation, the authors believe that experimentative materials will contribute to consistent embedding of CLIL courses into a complete system of vocational training.

Investigation of LBM-processed bimodal powder mixtures of the nickel base alloy HX and WC–Co

Laser beam melting (LBM) is an additive manufacturing technology (AM), which enables the production of individual, complex metal parts. The development of AM-specific materials is a necessary step to exploit the full technological potential and should be a useful contribution for the ongoing process of implementing, establishing, and realizing AM processes in industrial fields of application. Theoretically, the LBM process offers the opportunity to process any weldable metal powder feedstock. Since LBM parameters affect the solidification conditions directly, this process is predestinated for future material developments. In the present paper, first research results to process the conventional LBM powder (Hastelloy X; HX) that was blended with a fine WC–Co powder are presented. Due to its high temperature and corrosion resistance, HX is widely used in industrial AM applications. WC–Co was added to improve the wear and mechanical properties. In this study, the blended, bimodal powder feedstock was analyzed, and the influence of the key process parameters on the porosity, morphology, and grain structure was investigated. The aim was to obtain the first basic information concerning the behavior of such powder blends during LBM processing and the resulting material properties. In summary, processing powder blends is possible. Since an increased porosity was determined for a high WC–Co content, the process parameters need to be further improved. In addition, it was determined that the energy input during the powder melting process directly affects the distribution of WC–Co particles within the HX matrix. High-energy inputs lead to a degradation of the WC particles. Low-energy inputs foster a consistent embedding of WC–Co particles within the HX matrix.

Framework for an asymptotically safe standard model via dynamical breaking

We present a consistent embedding of the matter and gauge content of the Standard Model into an underlying asymptotically-safe theory, that has a well-determined in- teracting UV fixed point in the large colour/flavour limit. The scales of symmetry breaking are determined by two mass-squared parameters with the breaking of elec- troweak symmetry being driven radiatively. There are no other free parameters in the theory apart from gauge couplings.

Sparse Codes Auto-Extractor for Classification: A Joint Embedding and Dictionary Learning Framework for Representation

In this paper, we discuss the sparse codes auto-extractor based classification. A joint label consistent embedding and dictionary learning approach is proposed for delivering a linear sparse codes auto-extractor and a multi-class classifier by simultaneously minimizing the sparse reconstruction, discriminative sparse-code, code approximation and classification errors. The auto-extractor is characterized with a projection that bridges signals with sparse codes by learning special features from input signals for characterizing sparse codes. The classifier is trained based on extracted sparse codes directly. In our setting, the performance of the classifier depends on the discriminability of sparse codes, and the representation power of the extractor depends on the discriminability of input sparse codes, so we incorporate label information into the dictionary learning to enhance the discriminability of sparse codes. So, for inductive classification, our model forms an integration process from test signals to sparse codes and finally to assigned labels, which is essentially different from existing sparse coding based approaches that involve an extra sparse reconstruction with the trained dictionary for each test signal. Remarkable results are obtained by our model compared with other state-of-the-arts.

LeSSS: Learned Shared Semantic Spaces for Relating Multi‐Modal Representations of 3D Shapes

AbstractIn this paper, we propose a new method for structuring multi‐modal representations of shapes according to semantic relations. We learn a metric that links semantically similar objects represented in different modalities. First, 3D‐shapes are associated with textual labels by learning how textual attributes are related to the observed geometry. Correlations between similar labels are captured by simultaneously embedding labels and shape descriptors into a common latent space in which an inner product corresponds to similarity. The mapping is learned robustly by optimizing a rank‐based loss function under a sparseness prior for the spectrum of the matrix of all classifiers. Second, we extend this framework towards relating multi‐modal representations of the geometric objects. The key idea is that weak cues from shared human labels are sufficient to obtain a consistent embedding of related objects even though their representations are not directly comparable. We evaluate our method against common base‐line approaches, investigate the influence of different geometric descriptors, and demonstrate a prototypical multi‐modal browser that relates 3D‐objects with text, photographs, and 2D line sketches.