Search

Hypergraphs can model higher-order relationships among data objects that are found in applications such as social networks and bioinformatics. However, recent studies on hypergraph learning that extend graph convolutional networks to hypergraphs cannot learn effectively from features of unlabeled data. To such learning, we propose a contrastive hypergraph neural network, CHGNN, that exploits self-supervised contrastive learning techniques to learn from labeled and unlabeled data. First, CHGNN includes an adaptive hypergraph view generator that adopts an auto-augmentation strategy and learns a perturbed probability distribution of minimal sufficient views. Second, CHGNN encompasses an improved hypergraph encoder that considers hyperedge homogeneity to fuse information effectively. Third, CHGNN is equipped with a joint loss function that combines a similarity loss for the view generator, a node classification loss, and a hyperedge homogeneity loss to inject supervision signals. It also includes basic and cross-validation contrastive losses, associated with an enhanced contrastive loss training process. Experimental results on nine real datasets offer insight into the effectiveness of CHGNN, showing that it outperforms 19 competitors in terms of classification accuracy consistently.

We present a machine learning (ML) approach to static contact angle measurement, trained on a large data set (>7.2 million) of half drop contours based on solutions to the Young–Laplace equation where the contact angle is known a priori(removing all sources of error from human input). The data set included the effects of surface roughness, gravity, the size of drop relative to the image, and reflections of the drop on the surface. The presented ML model (valid for contact angles >110°), in combination with a new automated image and contour processing approach, is shown to be more accurate than other methods when benchmarked against an experimental data set, with an estimated error of 1°. The ML model is also 2 orders of magnitude faster at predicting contact angles than Young–Laplace fitting (the current best practice approach). The accuracy and speed of the presented approach provides a viable pathway toward robust and reproducible high-throughput contact angle analysis. This approach, Conan-ML, is open-source and provided for the use and development of new approaches to goniometry.

The task of entity alignment between knowledge graphs (KGs) aims to identify every pair of entities from two different KGs that represent the same entity. Many machine learning-based methods have been proposed for this task. However, to our best knowledge, existing methods all require manually crafted seed alignments, which are expensive to obtain. In this paper, we propose the first fully automatic alignment method named AutoAlign, which does not require any manually crafted seed alignments. Specifically, for predicate embeddings, AutoAlign constructs a predicate-proximity-graph with the help of large language models to automatically capture the similarity between predicates across two KGs. For entity embeddings, AutoAlign first computes the entity embeddings of each KG independently using TransE, and then shifts the two KGs’ entity embeddings into the same vector space by computing the similarity between entities based on their attributes. Thus, both predicate alignment and entity alignment can be done without manually crafted seed alignments. AutoAlign is not only fully automatic, but also highly effective. Experiments using real-world KGs show that AutoAlign improves the performance of entity alignment significantly compared to state-of-the-art methods.

The massive amount of data and large variety of data distributions in the big data era call for access methods that are efficient in both query processing and index bulk-loading, and over both practical and worst-case workloads. To address this need, we revisit a classic multidimensional access method - the R-tree. We propose a novel R-tree packing strategy that produces R-trees with an asymptotically optimal I/O complexity for window queries in the worst case. Our experiments show that the R-trees produced by the proposed strategy are highly efficient on real and synthetic data of different distributions. The proposed strategy is also simple to parallelize, since it relies only on sorting. We propose a parallel algorithm for R-tree bulk-loading based on the proposed packing strategy, and analyze its performance under the massively parallel communication model. Experimental results confirm the efficiency and scalability of the parallel algorithm over large data sets.

We study neural data-to-text generation to generate a sentence to describe a target entity based on its attributes. Specifically, we address two problems of the encoder-decoder framework for data-to-text generation: i) how to encode a non-linear input (e.g., a set of attributes); and ii) how to order the attributes in the generated description. Existing studies focus on the encoding problem but do not address the ordering problem, i.e., they learn the content-planning implicitly. The other approaches focus on two-stage models but overlook the encoding problem. To address the two problems at once, we propose a model named TransCP to explicitly learn content-planning and integrate them into a description generation model in an end-to-end fashion. We propose a novel Transformer-based Pointer Network with gated residual attention and importance masking to learn a content-plan. To integrate the content-plan with a description generator, we propose a tracking mechanism to trace the extent to which the content-plan is exposed in the previous decoding time-step. This helps the description generator select the attributes to be mentioned in proper order. Experimental results show that our model consistently outperforms state-of-the-art baselines by up to 2% and 3% in terms of BLEU score on two real-world datasets.

Region similarity learning plays an essential role in applications such as business site selection, region recommendation, and urban planning. Earlier studies mainly represent regions as bags of points of interest (POIs) for region similarity comparisons, which cannot fully exploit the spatial features of the regions. Recently, researchers propose to use deep neural networks to exploit spatial features such as POI geo-coordinates and categories, which have produced more accurate and robust region similarity learning results. However, many useful features such as the height and size of a POI, and the distance and relative importance between the POIs, are still overlooked in these methods. To take advantage of such features, we propose to represent regions as graphs, where nodes are POIs with rich features such as height, size, and hexagonal coordinates, while edges are the relationships between POIs formulated by their road network distances. To capture POIs’ importance, we weigh them by their height and size. Since there is limited availability of ground-truth region similarity data, we propose a contrastive learning-based multi-relational graph neural network (C-MPGCN) for region similarity learning based on the graph representations. To generate data for model training, we propose a soft graph edit distance (SGED) based algorithm to generate triples of similar and dissimilar graphs of a given graph (representing a given region) based on the POI weights. Experimental results show that C-MPGCN outperforms the state-of-the-art methods for region similarity learning consistently with an improvement of at least 8.6% and 9.4% in terms of MRR and HR@1, respectively.

Designed ligands of G protein-coupled receptors can exert a spectrum of modulating effects, varying from full agonists and partial agonists to antagonists and inverse agonists. For the dopamine D2receptor (D2R), partial agonist activity is the pharmacological feature of the third-generation antipsychotics, including aripiprazole, brexpiprazole, and cariprazine. Started from a benzofuran-derived D2R full agonist O4LE6(4), which was identified using a structure-based method by us in previous studies, a series of D2R partial agonists were designed and synthesized by introducing different tail groups. Among them, compound 10bshowed excellent activity in D2R pharmacological assays. Further optimizations using a structural rigidification approach led to the discovery of brain-penetrant compounds 29cand 29d, which exhibited potent antipsychotic effects in the mouse hyperlocomotion model. Compound 29calso showed excellent drug-like pharmacokinetic properties in rats and qualifies as an antipsychotic agent that is worth further evaluations.

Dispersed computing is a new resource-centric computing paradigm. Due to its high degree of openness and decentralization, it is vulnerable to attacks, and security issues have become an important challenge hindering its development. The trust evaluation technology is of great significance to the reliable operation and security assurance of dispersed computing networks. In this paper, a dynamic Bayesian-based comprehensive trust evaluation model is proposed for dispersed computing environment. Specifically, in the calculation of direct trust, a logarithmic decay function and a sliding window are introduced to improve the timeliness. In the calculation of indirect trust, a random screening method based on sine function is designed, which excludes malicious nodes providing false reports and multiple malicious nodes colluding attacks. Finally, the comprehensive trust value is dynamically updated based on historical interactions, current interactions and momentary changes. Simulation experiments are introduced to verify the performance of the model. Compared with existing model, the proposed trust evaluation model performs better in terms of the detection rate of malicious nodes, the interaction success rate, and the computational cost.

Traffic prediction is an important and yet highly challenging problem due to the complexity and constantly changing nature of traffic systems. To address the challenges, we propose a graph and attentive multi-path convolutional network (GAMCN) model to predict traffic conditions such as traffic speed across a given road network into the future. Our model focuses on the spatial and temporal factors that impact traffic conditions. To model the spatial factors, we propose a variant of the graph convolutional network (GCN) named LPGCN to embed road network graph vertices into a latent space, where vertices with correlated traffic conditions are close to each other. To model the temporal factors, we use a multi-path convolutional neural network (CNN) to learn the joint impact of different combinations of past traffic conditions on the future traffic conditions. Such a joint impact is further modulated by an attention generated from an embedding of the prediction time, which encodes the periodic patterns of traffic conditions. We evaluate our model on real-world road networks and traffic data. The experimental results show that our model outperforms state-of-art traffic prediction models by up to 18.9% in terms of prediction errors and 23.4% in terms of prediction efficiency.

Partial agonist activity at the dopamine D2receptor (DRD2) is a key feature of third-generation antipsychotics (TGAs). However, TGAs also act as antagonists or weak partial agonists to the serotonin (5-hydroxytryptamine; 5-HT) 2A receptor (5-HT2AR). Here we present the crystal structures of aripiprazole- and cariprazine-bound human 5-HT2AR. Both TGAs adopt an unexpected ‘upside-down’ pose in the 5-HT2AR binding pocket, with secondary pharmacophores inserted in a similar way to a ‘bolt’. This insight into the binding modes of TGAs offered a structural mechanism underlying their varied partial efficacies at 5-HT2AR and DRD2. These structures enabled the design of a partial agonist at DRD2/3 and 5-HT1AR with negligible 5-HT2AR binding that displayed potent antipsychotic-like activity without motor side effects in mice. This TGA lead also had antidepressant-like effects and improved cognitive performance in mouse models via 5-HT1AR. This work indicates that 5-HT2AR affinity is a dispensable contributor to the therapeutic actions of TGAs.

The massive amount of data and large variety of data distributions in the big data era call for access methods that are efficient in both query processing and index bulk-loading, and over both practical and worst-case workloads. To address this need, we revisit a classic multidimensional access method - the R-tree. We propose a novel R-tree packing strategy that produces R-trees with an asymptotically optimal I/O complexity for window queries in the worst case. Our experiments show that the R-trees produced by the proposed strategy are highly efficient on real and synthetic data of different distributions. The proposed strategy is also simple to parallelize, since it relies only on sorting. We propose a parallel algorithm for R-tree bulk-loading based on the proposed packing strategy, and analyze its performance under the massively parallel communication model. Experimental results confirm the efficiency and scalability of the parallel algorithm over large data sets.

Using solar energy to produce syngas via the endothermic reforming of methane has been extensively investigated at the laboratory- and pilot plant-scales as a promising method of storing solar energy. One of the challenges to scaling up this process in a tubular reformer is to improve the reactor's performance, which is limited by mass and heat transfer issues. High thermal conductivity Cu foam was therefore used as a substrate to improve the catalyst's thermal conductivity during solar reforming. We also developed a method to coat the foam with the catalytically active component NiMg3AlOx. The Cu foam-based NiMg3AlOxperforms better than catalysts supported on SiSiC foam, which is currently used as a substrate for solar-reforming catalysts, at high gas hourly space velocity (≥400,000 mL/(g·h)) or at low reaction temperatures (≤ 720 °C). The presence of a γ-Al2O3intermediate layer improves the adhesion between the catalyst and substrate as well as the catalytic activity.

The purpose of this paper is to investigate heat dissipation performance of porous copper with long cylindrical pores fabricated by a unidirectional solidification method. Three samples with porosity of 29.87%, 34.47% and 50.98% were chosen and cut into size of 60 mm (length) × 26 mm (width) × 2 mm (thickness) along the vertical direction of pore axis. Their heat dissipation performance was evaluated by a nonsteady method in air and compared to those of not only bulk copper but also bored coppers with porosity of 30.61% and 32.20%. It is found that the porous copper dissipated heat faster by a forced air convection than that by natural convection from 80 °C to room temperature and both porosity and pore size play an important role in the performance for the porous copper. Furthermore, the heat dissipation rate is higher when the forced air was circulated along the specimens than that perpendicular to the specimens for the porous copper. It is revealed that porous copper with bigger porosity and a proper pore size possesses a higher heat dissipation rate. It is concluded that the porous copper with elongated cylindrical pores has larger heat dissipation performance than both the bulk copper and the bored copper, which is attributed to its higher specific surface area. Application of the porous copper for heat dissipation is promising. [ABSTRACT FROM AUTHOR]

In this paper, the bit synchronization algorithms in GNSS receiver are introduced, including the traditional histogram method, K-P algorithm and Viterbi algorithm. The FPGA implementation is also included. A novel time division multiplexing technology (TDM) based on multi-channel shared synchronizer is proposed in this paper to solve the constrained hardware resource problem of multi-system satellite navigation receiver. Through the using of control state machine and data register structure, we realize the multiplexing of bit synchronizer of navigation receiver, which saves the hardware resource. After the experiment, it can be verified that the receiver based on the bit synchronization and multiplexing technology can correctly restore the navigation information.