Single Representation Research Articles

INCEPT: A Framework for Duplicate Posts Classification with Combined Text Representations

Dealing with many of the problems related to the quality of textual content online involves identifying similar content. Algorithmic solutions for duplicate content classification typically rely on text vector representation, which maps textual information into a set of features. Ideally, this representation would capture all aspects of the underlying text, including length, word frequencies, syntax, and semantics. While recent advancements in text representation have led to improved performance, a comprehensive approach that explicitly incorporates all text features has not yet been proposed. In this study, we present the INCEPT framework that utilizes multiple representation methods to detect duplicate text pairs, taking advantage of their individual strengths. The core of our approach involves using a stacking ensemble of pairwise vector distance measurements that are computed from multiple text representation methods. A stacking classifier then utilizes these distance scores as input and learns to identify duplicate posts. We assess the proposed framework’s effectiveness in identifying duplicate posts in an online Question and Answer platform. By combining several text representation methods, INCEPT performs well in the duplicate posts classification task. Our experiments demonstrate that specific framework configurations outperform the accuracy scores obtained from individual text representation methods. Therefore, we also infer that no single text representation method can independently capture a text’s features.

HGBER: Heterogeneous Graph Neural Network With Bidirectional Encoding Representation.

Heterogeneous graphs with multiple types of nodes and link relationships are ubiquitous in many real-world applications. Heterogeneous graph neural networks (HGNNs) as an efficient technique have shown superior capacity of dealing with heterogeneous graphs. Existing HGNNs usually define multiple meta-paths in a heterogeneous graph to capture the composite relations and guide neighbor selection. However, these models only consider the simple relationships (i.e., concatenation or linear superposition) between different meta-paths, ignoring more general or complex relationships. In this article, we propose a novel unsupervised framework termed Heterogeneous Graph neural network with bidirectional encoding representation (HGBER) to learn comprehensive node representations. Specifically, the contrastive forward encoding is firstly performed to extract node representations on a set of meta-specific graphs corresponding to meta-paths. We then introduce the reversed encoding for the degradation process from the final node representations to each single meta-specific node representations. Moreover, to learn structure-preserving node representations, we further utilize a self-training module to discover the optimal node distribution through iterative optimization. Extensive experiments on five open public datasets show that the proposed HGBER model outperforms the state-of-the-art HGNNs baselines by 0.8%-8.4% in terms of accuracy on most datasets in various downstream tasks.

Development of a multi-layer canopy model for E3SM Land Model with support for heterogeneous computing

The vertical structure of vegetation canopies creates micro-climates. However, the land components of most Earth System Models, including the Energy Exascale Earth System Model (E3SM), typically neglect vertical canopy structure by using a single layer big-leaf representation to simulate water, CO2, and energy exchanges between the land and the atmosphere. In this study, we developed a Multi-Layer Canopy Model for the E3SM Land Model to resolve the micro-climate created by vegetation canopies. The model developed in this study re-implements the CLM-ml_v1 to support heterogeneous computing architectures consisting of CPUs and GPUs and includes three additional optimization-based stomatal conductance models. The use of Portable, Extensible Toolkit for Scientific Computation provides a speedup of 25–50 times on a GPU relative to a CPU. The numerical implementation of the model was verified against CLM-ml_v1 for a month-long simulation using data from the Ameriflux US-University of Michigan Biological Station site. Model structural uncertainty was explored by performing control simulations for five stomatal conductance models that exclude and include the control of plant hydrodynamics (PHD) on photosynthesis. The bias in simulated sensible and latent heat fluxes was lower when PHD was accounted for in the model. Additionally, six idealized simulations were performed to study the impact of three environmental variables (i.e. air temperature, atmospheric CO2, and soil moisture) on canopy processes (i.e. net CO2 assimilation, leaf temperature, and leaf water potential). Increasing air temperature reduced net CO2 assimilation and increased air temperature. Net CO2 assimilation increased at higher atmospheric CO2, while decreasing soil moisture resulted in lower leaf water potential.

H-SLAM: Hybrid direct–indirect visual SLAM

The recent success of hybrid methods in monocular odometry has led to many attempts to generalize the performance gains to hybrid monocular SLAM. However, most attempts fall short in several respects, with the most prominent issue being the need for two different map representations (local and global maps), with each requiring different, computationally expensive, and often redundant processes to maintain. Moreover, these maps tend to drift with respect to each other, resulting in contradicting pose and scene estimates, and leading to catastrophic failure. In this paper, we propose a novel approach that makes use of descriptor sharing to generate a single inverse depth scene representation. This representation can be used locally, queried globally to perform loop closure, and has the ability to re-activate previously observed map points after redundant points are marginalized from the local map, eliminating the need for separate map maintenance processes. The maps generated by our method exhibit no drift between each other, and can be computed at a fraction of the computational cost and memory footprint required by other monocular SLAM systems. Despite the reduced resource requirements, the proposed approach maintains its robustness and accuracy, delivering performance comparable to state-of-the-art SLAM methods (e.g., LDSO, ORB-SLAM3) on the majority of sequences from well-known datasets like EuRoC, KITTI, and TUM VI. The source code is available at: https://github.com/AUBVRL/fslam_ros_docker.

Vector representations and unit vector representations of fields: Problems of understanding and possible teaching strategies

Vector fields are a highly abstract physical concept that is often taught using visualizations. Although vector representations are particularly suitable for visualizing quantitative data, they are often confusing, especially when describing real fields such as magnetic and electric fields, as the vector arrows can overlap. The present study investigates vector understanding at the end of secondary education. In particular, the extent to which the geometry of the field can be derived from conventional unit vector representations and representations with centered unit vectors was examined. To support this understanding, two exercises were compared. The unirepresentational exercise argued within the conventional unit vector representation, while the multirepresentational exercise attempted to support the link between centered and conventional unit vectors. The results show that almost all test subjects solved the items for generating vector representations correctly, but significant difficulties were encountered in interpreting vector representations. Drawing and interpreting vector representations therefore appear to be different skills that should be practiced intensively and in an integrated way. Various problems could be identified when interpreting vector representations. For example, the number of vectors is often erroneously used to estimate the strength of the field, although more vectors per surface element actually only increase the resolution of the representation. Here, however, the results suggest that the longitudinal density and the transverse density of the drawn vectors are perceived differently by the learners. Furthermore, the learners recognized the field’s geometry much more readily from centered unit vectors than from conventional unit vectors. Errors occur especially when interpreting the geometry of conventional unit vector representations of rotational fields and fields containing both sources and sinks while the geometries of fields containing only sinks were interpreted quite well. The comparison between the two training exercises showed that a promising approach to deepen students’ understanding would be to use an exercise that contrasts conventional and centered unit vector representations and explains how to translate from one representation to the other, rather than describing the main elements of only a single representation. Finally, based on the results of the study, we propose a strategy for teaching vector representations in schools. Given the significantly improved readability of the representation with centered unit vectors, the results even raise the question of whether this type of representation could possibly replace the conventional representation in textbooks and learning materials in the future. Published by the American Physical Society 2024

Analysis of Ambiguity, Vagueness, Fuzziness, Uncertainty, Possibility and Probability in the Natural Language Semantics with Fuzzy Logic

Understanding the esotericism of human instinct in their daily life conversation is not enough then a mystery now. This is a bundle of ambiguity, vagueness, fuzziness, uncertainty, possibility and probability as a wrap that humans have built around themselves. With the advancement in artificial Intelligence, natural language processing is more capable now to work with real world and performing intelligent analysises. The real world has interactions between natural and artificial intelligent systems. Despite all it, humans retained their superiority over artificial intelligent systems. The fuzzy Logic can play an important computational role in understanding this intelligence gap in clear dimensions. Logical Semantics, Distributional Semantics and Probabilistic Logic are focused on their intention for better natural language semantic representations. But no single semantic representation fulfills all requirements needed for a satisfactory representation. The objective of the present work has two folds. The first one focused on the understanding of fuzzy logic in two dimensions as an intelligence computational technique and another as mathematical modeling of natural language semantics. The second fold illustrates this intelligence gap with real world examples of natural language processing applications such as Google and Microsoft Translator.

MvMRL: a multi-view molecular representation learning method for molecular property prediction.

Effective molecular representation learning is very important for Artificial Intelligence-driven Drug Design because it affects the accuracy and efficiency of molecular property prediction and other molecular modeling relevant tasks. However, previous molecular representation learning studies often suffer from limitations, such as over-reliance on a single molecular representation, failure to fully capture both local and global information in molecular structure, and ineffective integration of multiscale features from different molecular representations. These limitations restrict the complete and accurate representation of molecular structure and properties, ultimately impacting the accuracy of predicting molecular properties. To this end, we propose a novel multi-view molecular representation learning method called MvMRL, which can incorporate feature information from multiple molecular representations and capture both local and global information from different views well, thus improving molecular property prediction. Specifically, MvMRL consists of four parts: a multiscale CNN-SE Simplified Molecular Input Line Entry System (SMILES) learning component and a multiscale Graph Neural Network encoder to extract local feature information and global feature information from the SMILES view and the molecular graph view, respectively; a Multi-Layer Perceptron network to capture complex non-linear relationship features from the molecular fingerprint view; and a dual cross-attention component to fuse feature information on the multi-views deeply for predicting molecular properties. We evaluate the performance of MvMRL on 11 benchmark datasets, and experimental results show that MvMRL outperforms state-of-the-art methods, indicating its rationality and effectiveness in molecular property prediction. The source code of MvMRL was released in https://github.com/jedison-github/MvMRL.

Hierarchical multimodal self-attention-based graph neural network for DTI prediction.

Drug-target interactions (DTIs) are a key part of drug development process and their accurate and efficient prediction can significantly boost development efficiency and reduce development time. Recent years have witnessed the rapid advancement of deep learning, resulting in an abundance of deep learning-based models for DTI prediction. However, most of these models used a single representation of drugs and proteins, making it difficult to comprehensively represent their characteristics. Multimodal data fusion can effectively compensate for the limitations of single-modal data. However, existing multimodal models for DTI prediction do not take into account both intra- and inter-modal interactions simultaneously, resulting in limited presentation capabilities of fused features and a reduction in DTI prediction accuracy. A hierarchical multimodal self-attention-based graph neural network for DTI prediction, called HMSA-DTI, is proposed to address multimodal feature fusion. Our proposed HMSA-DTI takes drug SMILES, drug molecular graphs, protein sequences and protein 2-mer sequences as inputs, and utilizes a hierarchical multimodal self-attention mechanism to achieve deep fusion of multimodal features of drugs and proteins, enabling the capture of intra- and inter-modal interactions between drugs and proteins. It is demonstrated that our proposed HMSA-DTI has significant advantages over other baseline methods on multiple evaluation metrics across five benchmark datasets.

Topological 3D reconstruction of multiple anatomical structures from volumetric medical data

In the medical field, usually, practitioners mainly base their analysis on 2D slices produced from MRI or CT-scans that correspond to restricted views of a pathology. To facilitate the work of doctors, increase diagnostic accuracy and cross-reference multimodal data, a 3D reconstruction is required. However, most of the time, reconstruction methods fail at visualizing complex and noisy data made up of several tissues. Indeed, these methods often build each tissue independently so that the consistency of the global model is not ensured: overlaps may appear between segments whereas some disjointed volumes exhibit empty spaces. This paper presents a complete topologically consistent reconstruction system from 3D medical acquisitions such as MRI or CT-scans. Compared to other methods, our system offers a single volumetric representation of an organ corresponding to a 3D space partition, where a semantic label is associated to each volume to identify the represented tissue and adjacency between volumes is explicitly and precisely defined. This partition is controlled and free from topological and geometric defects usually found in other 3D reconstruction approaches. Experimental studies were conducted on MRI datasets of brains resulting in consistent reconstructions. An application of the model for calculating the distribution of physiological data in brain tissue is also shown.

A Multi-view Molecular Pre-training with Generative Contrastive Learning.

Molecular representation learning can preserve meaningful molecular structures as embedding vectors, which is a necessary prerequisite for molecular property prediction. Yet, learning how to accurately represent molecules remains challenging. Previous approaches to learning molecular representations in an end-to-end manner potentially suffered information loss while neglecting the utilization of molecular generative representations. To obtain rich molecular feature information, the pre-training molecular representation model utilized different molecular representations to reduce information loss caused by a single molecular representation. Therefore, we provide the MVGC, a unique multi-view generative contrastive learning pre-training model. Our pre-training framework specifically acquires knowledge of three fundamental feature representations of molecules and effectively integrates them to predict molecular properties on benchmark datasets. Comprehensive experiments on seven classification tasks and three regression tasks demonstrate that our proposed MVGC model surpasses the majority of state-of-the-art approaches. Moreover, we explore the potential of the MVGC model to learn the representation of molecules with chemical significance.

Profile of Pre-service Physics Teachers' Representational Fluency on Electrostatic Concept

Science teachers have to master representational to communicate and be aware of the students’ difficulties in understanding science concepts. This study aims to determine representational fluency as part of science communication skills in pre-service physics teachers. This study uses descriptive analysis techniques based on the percentage. The research subjects are 50 pre-service physics teachers aged between 19 and 22 years. This study was conducted at the Study Program of Physics Education of a college in Maluku. How fluent is a pre-service physics teacher in representing the concept of electrostatic was measured using 15 valid and reliable representational fluency multiple choice test that includes four component of representational fluency: constructing single representation, constructing multiple representation, translating between representation and reviewing single representation. The findings of the present study indicate that although students had started to learn concepts of electrostatic their representational fluency is still low. The preservice teachers’ rate of giving correct answers to the test items varies between 8% and 48%. The mean score of the pre-service teachers was found to be 4.06. Keywords: representational fluency, electrostatic

Multimodal protein representation learning and target-aware variational auto-encoders for protein-binding ligand generation

Without knowledge of specific pockets, generating ligands based on the global structure of a protein target plays a crucial role in drug discovery as it helps reduce the search space for potential drug-like candidates in the pipeline. However, contemporary methods require optimizing tailored networks for each protein, which is arduous and costly. To address this issue, we introduce TargetVAE, a target-aware variational auto-encoder that generates ligands with desirable properties including high binding affinity and high synthesizability to arbitrary target proteins, guided by a multimodal deep neural network built based on geometric and sequence models, named Protein Multimodal Network (PMN), as the prior for the generative model. PMN unifies different representations of proteins (e.g. primary structure—sequence of amino acids, 3D tertiary structure, and residue-level graph) into a single representation. Our multimodal architecture learns from the entire protein structure and is able to capture their sequential, topological, and geometrical information by utilizing language modeling, graph neural networks, and geometric deep learning. We showcase the superiority of our approach by conducting extensive experiments and evaluations, including predicting protein-ligand binding affinity in the PBDBind v2020 dataset as well as the assessment of generative model quality, ligand generation for unseen targets, and docking score computation. Empirical results demonstrate the promising and competitive performance of our proposed approach. Our software package is publicly available at https://github.com/HySonLab/Ligand_Generation.

Grouping in working memory guides chunk formation in long-term memory: Evidence from the Hebb effect

The Hebb effect refers to the improvement in immediate memory performance on a repeated list compared to unrepeated lists. That is, participants create a long-term memory representation over repetitions, on which they can draw in working memory tests. These long-term memory representations are likely formed by chunk acquisition: The whole list becomes integrated into a single unified representation. Previous research suggests that the formation of such chunks is rather inflexible and only occurs when at least the beginning of the list repeats across trials. However, recent work has shown that repetition learning strongly depends on participants recognizing the repeated information. Hence, successful chunk formation may depend on the recognizability of the repeated part of a list, and not on its position in the list. Across six experiments, we compared these two alternatives. We tested immediate serial recall of eight-letter lists, some of which partially repeated across trials. We used different partial-repetition structures, such as repeating only the first half of a list, or only every second item. We manipulated the salience of the repeating structure by spatially grouping and coloring the lists according to the repetition structure. We found that chunk formation is more flexible than previously assumed: Participants learned contiguous repeated sequences regardless of their position within the list, as long as they were able to recognize the repeated structure. Even when the repeated sequence occurred at varying positions over repetitions, learning was preserved when the repeated sequence was made salient by the spatial grouping. These findings suggest that chunk formation requires recognition of which items constitute a repeating group, and demonstrate a close link between grouping of information in working memory, and chunk formation in long-term memory.

3D hand pose estimation and reconstruction based on multi-feature fusion

3D hand pose estimation and shape reconstruction is to recover the hand joint points and hand mesh vertices coordinates from the image. However, existing methods usually only use the high-level semantic features extracted by the backbone network to represent the hand mesh vertex features, which leads to a single representation of the hand vertices features and cannot fully utilize the feature information extracted by the network. In this paper, we propose a method for real-time 3D reconstruction of hands from a single RGB image, which enriches the 3D semantic information of the mesh vertices through multi-feature fusion. Firstly, we regress the 2D features of mesh vertices through Integral Pose Regression (IPR) and regard them as prior information to 3D features. Then we design a Multi-Scale Sampling(MSS) module to extract multi-scale information. Finally we fuse 2D prior features, multi-scale features, and high-level semantic features extracted by backbone to represent 3D initial feature. Additionally, we propose a Multi-Root(MR) loss function to address the imbalance problem caused by a single root joint. The experimental results indicate that our network achieves competitive performance on the FreiHAND and HO-3D public datasets, achieving fast inference speed with fewer parameters.

Multi-modal fusion network guided by prior knowledge for 3D CAD model recognition

Currently, the 3D CAD model has found extensive use in various applications. The more accurate recognition and more reliable reuse of 3D computer-aided design (CAD) models could significantly save labor costs and time in the production and design of 3D products. However, existing 3D model recognition methods are constrained by a single representation form that lacks supplementary information from multiple sources, and there is a lack of corresponding guidance for relationships between different data. In this paper, we focus on multi-modal information for 3D model representation and propose a Multi-modal Fusion Network (MMFN) guided by prior knowledge for 3D CAD model recognition. In particular, we consider prior knowledge about the class distribution of 3D models: The designed labeling information is utilized as the prior knowledge to guide the multi-modal information fusion via the cross-attention structure. Then, the contrastive learning method is utilized in the optimization step, further increasing the aggregation of similar samples and contributing to enhancing the discriminative capability of features. Finally, we conduct extensive experiments on the public datasets, ModelNet, ShapeNet, and a challenging industrial 3D CAD dataset built by ourselves. Compared to state-of-the-art approaches, our MMFN provides competitive results. The source code has been published on Github: https://github.com/WhiteTJU/MMFN.

Assessing input parameter hyperspace and parameter identifiability in a cardiovascular system model via sensitivity analysis

We aim to clarify our understanding of the process of state-space model input parameter identification, known, within the clinical context, as model personalisation. To do so, we apply reference sensitivity and identifiability techniques to a lumped parameter, single ventricle representation of the systemic circulation, chosen in view of its relative simplicity and prior art. We attempt to quantify the reliability of input parameter identifiability through the lens of 4 clinically relevant measurements and the attendant difficulty in personalising the model. In turn, this that we extend existing methods which combine both parameter influence and orthogonality, to global sensitivities. By examining different parameter sensitivity evaluation methodologies, we investigate the stability of optimal parameter subsets which are commonly used to aid clinical investigations. In order to perform the personalisation process, one must understand the complexity of the high dimensional input parameter hyperspace associated with this class of model. By utilising Sobol indices, we propose a domain-agnostic and intuitive approach. This involves varying the bounds of the input parameter space relative to the model’s base state. These investigations yield a pseudo-mapping of the input hyperspace, cementing our understanding of the role of identifiable input parameters in the state-space model. Our findings suggest a novel global methodology for input parameter identifiability and input hyperspace mapping, providing valuable insights into solving the personalisation process.

Rock engravings marking the Redan landscape in Gauteng, South Africa

The ontological1 and epistemological2 foundations of reality, myths, beliefs, and practices in ritual behaviour have long been studied by scholars around the globe. There is no single understanding and representation of reality, as all that is real or existent within the universe is historically shaped by interdependent cultural and social processes and fear factors. Although the study of rock art is primary a branch of archaeology, cultural history also focuses on a past time and place, to study beliefs and ideas from a bottom-up approach and include the objects and experiences (sometimes unwritten) of the everyday life of the uneducated and less privileged. It is particularly well equipped to offer an understanding of the San myths, beliefs, ritual practices, and modes of representation of reality, as represented in their rock art. The Redan open-air rock engraving site is located on the Kookfontein farm, near the former Klip power station at Redan, close to a wetland that was formerly a rivulet that founded its way to the Klip River, a tributary of the Vaal River. The Klip-Vaal area is situated in the southern part of Gauteng, South Africa. Being part of the Late Stone Age hunting, domestic and ritual sites of the San people and later migrants it contains numerous documented and undocumented rock art sites. In line with interpretivism, the methodology is based on published and unpublished literature, as well as several on-site inspections of the site between 2002 and 2022. Authorship of the engravings at the Redan site is unconfirmed, although two dominant views regarding the cultural origin have emerged, namely a San hunter-gatherer shamanistic approach and the Koranna-Khoekhoen initiation site hypothesis.Contribution: The article contributes to the neglected historiography of the rock engravings in the Klip-Vaal region, in particular the Redan site, and the need to facilitate a comparison with similar rock engravings in southern Africa.

What makes different number-space mappings interact?

Models of numerical cognition consider a visuo-spatial representation to be at the core of numerical processing, the 'mental number line'. Two main interference effects between number and space have been described: the SNARC effect reflects a small number/left side and large number/right side association (number-location mapping); the size-congruity effect (SCE) reflects a small number/small size and large number/large size association (number-size mapping). Critically, a thorough investigation on the representational source for these two number-space mappings is lacking, leaving open the question of whether the same representation underlies both phenomena. Here, we build on a recent study (Viarouge and de Hevia in Front Hum Neurosci 15:750964, 2021) in order to address this question in three experiments, by systematically manipulating the presence of the two conditions that might elicit an interaction between SNARC and SCE: (i) an implicit task whereby numerical and spatial information are task-irrelevant, (ii) a design in which the number-space congruency relative to both mappings vary at the same level -either both within or between blocks. Experiment 1 replicated the interaction between the two mappings when both factors were present. Experiments 2 and 3 dissociated the two factors by varying the two mappings at the same level but using an explicit comparison task (Experiment 2), or by using an implicit task but with mappings varying at different levels (Experiment 3). We found that both factors, either in combination or used in isolation, drive the interaction between the two number-space mappings. These findings are discussed in terms of the weight given to each mapping, suggesting that a single representation encompassing both number-space mappings is therefore activated whenever both mappings are given equal weight through task requirements.

Blocked training facilitates learning of multiple schemas

We all possess a mental library of schemas that specify how different types of events unfold. How are these schemas acquired? A key challenge is that learning a new schema can catastrophically interfere with old knowledge. One solution to this dilemma is to use interleaved training to learn a single representation that accommodates all schemas. However, another class of models posits that catastrophic interference can be avoided by splitting off new representations when large prediction errors occur. A key differentiating prediction is that, according to splitting models, catastrophic interference can be prevented even under blocked training curricula. We conducted a series of semi-naturalistic experiments and simulations with Bayesian and neural network models to compare the predictions made by the “splitting” versus “non-splitting” hypotheses of schema learning. We found better performance in blocked compared to interleaved curricula, and explain these results using a Bayesian model that incorporates representational splitting in response to large prediction errors. In a follow-up experiment, we validated the model prediction that inserting blocked training early in learning leads to better learning performance than inserting blocked training later in learning. Our results suggest that different learning environments (i.e., curricula) play an important role in shaping schema composition.

A Multiinterest and Social Interest-Field Framework for Financial Security

Online payment has become an influential method of transaction. While improving convenience, this method of payment also brings great financial risks. Prevalidating transactions can be effective in reducing the number of frauds. For this reason, recommendation algorithms have been introduced to measure the credibility of transactions by predicting users’ ratings of items. However, most algorithms deal with the relationships in social networks without distinction, mixing positive and negative information into the recommender system, which brings huge noise. And, they only generate a single interest representation for each user to measure the similarity between users and spread interest, ignoring the diversity of user interests. Moreover, they did not consider the propagation of different interests would be different. In this article, we propose a multiinterest and social interest-field framework (MISIF) for social recommendations in financial security, which introduces capsule networks into social recommendation and extends the traditional single-interest representation to user multiinterest embedding by dynamic routing (DR) and other methods to improve the expressiveness of user embedding. After that, we construct social interest fields to integrate social interests based on multiinterest embedding, which alleviates the noise in social networks and user data sparsity problems. Finally, we aggregate user multiinterest embedding and additional information through neural networks to obtain the final prediction scores. Experiments with three publicly available datasets show that our proposed MISIF framework outperforms the state-of-the-art social recommendation methods.