The online‐to‐offline (O2O) business model has facilitated millions of daily transactions on popular online food ordering platforms. Online food delivery route planning presents a complex multidepot vehicle routing problem (VRP) with capacity limits, pickup–delivery, and time‐window constraints. However, the vast volume of transactions and computational complexities of delivery routes pose significant challenges. This paper proposes a novel feature fusion attention‐based deep reinforcement learning model to address such constrained routing problems. The innovative encoding and masking scheme with a self‐attention‐guided order relocation operator efficiently handles multidepot and multiconstraint scenarios. Additionally, incorporating self‐attention with a graph neural network (GNN) framework extends existing research from static unit square environments to dynamic real road networks. Computational experiments demonstrate that our proposed route solver outperforms state‐of‐the‐art heuristics and reinforcement learning methods regarding solution quality and computation time across unit square environments and real road networks. Exploratory analysis using real‐world delivery data a illustrates the applicability of our approach to practical online ordering platforms.

Drug repositioning has become a transformative paradigm in pharmaceutical research, with accurate prediction of drug-disease associations serving as its computational cornerstone. While graph convolutional networks have shown promising results in this domain, their architecture designed for homogeneous graphs fundamentally limits the effectiveness in modeling heterogeneous biological networks with diverse entity types and interactions, especially the propagation of collaborative signals across different entities. In this study, we propose HeteroNGCF, a network representation learning framework that effectively models high-order connectivity patterns in the drug-gene-proteindisease network while preserving the semantic heterogeneity of different entities. Specifically, HeteroNGCF firstly projects node features in homogeneous graphs into a unified latent space, enabling the alignment of heterogeneous entities while preserving their distinct semantic properties. To capture collaborative signals latent in heterogeneous interactions, we employ the neural graph collaborative filtering technique that integrates bipartite graph structures into the embedding process by recursively aggregating neighborhood information through layer-wise propagation. Afterwards, we aggregate embeddings learned from diverse interaction networks to obtain expressive representations of drugs and diseases, thereby enhancing prediction performance. Extensive experiments demonstrate that HeteroNGCF outperforms baseline methods across all evaluation metrics (AUC, AUPR, F1-score, recall and precision).

Recent Deep Reinforcement Learning (DRL) techniques have advanced solutions to Vehicle Routing Problems (VRPs). However, many of these methods focus exclusively on optimizing distance-oriented objectives (i.e., minimizing route length), often overlooking the implicit drivers' preferences for routes. These preferences, which are crucial in practice, are challenging to model using traditional DRL approaches. To address this gap, we propose a preference-based DRL method characterized by its reward design and optimization objective, which is specialized to learn historical route preferences. Our experiments demonstrate that the method aligns generated solutions more closely with human preferences. Moreover, it exhibits strong generalization performance across a variety of instances, offering a robust solution for different VRP scenarios.

Many real-world applications often generate unbalanced data. Learning from such data may lead to biased classifiers that perform poorly on the class of interest. Oversampling methods have been shown to be effective in rebalancing unbalanced data to help classifiers avoid performance bias. However, many existing oversampling methods rely on a predesigned linear model structure and the neighborhood information of an original instance. This may lead to the generation of noisy instances when the original data has noise. In this study, we develop a novel oversampling method in which genetic programming is introduced to automatically select good-quality instances and evolve a model structure that combines the selected instances to create a new instance. In the proposed oversampling method, an individual is used to represent a generated instance, which is evaluated by the fitness function designed based on the Euclidean distance and the cosine theorem. In the experiments, we examine the effectiveness of the proposed oversampling method in assisting different types of classifiers to solve the issue of class imbalance, and compare it with popular sampling methods in unbalanced classification. The results have been analyzed comprehensively, indicating that the new method successfully addressed the class imbalance issue by generating a group of good-quality instances for the minority class and outperformed the compared sampling methods in almost all cases.

As scientific communities become increasingly complex and interdisciplinary, implementing policies that foster collective creativity for science-based innovation emerges as a critical issue. In this paper, we investigate the impact of socio-communication strategies on innovation capacity based on experiments using a general-purpose model that aims to study dynamic processes of collective creativity in scientific communities. Experimental results indicate that low communication frequency, openness and receptivity lead to higher variety. On the contrary, variety decreases with increasing receptivity under high communication frequency. If more communication channels exist, the connectivity-driven preferential attachment strategy is beneficial in fostering innovation in terms of generated networks with low density and high centrality. Additionally, balance theory suggests higher levels of knowledge diffusion efficiency.

In order to visualize and analyze complex collective data, complicated geometric structure of each data is desired to be mapped onto a canonical domain to enable map-based visual exploration. This paper proposes a novel volume-preserving mapping and registration method which facilitates effective collective data visualization. Given two 3-manifolds with the same topology, there exists a mapping between them to preserve each local volume element. Starting from an initial mapping, a volume restoring diffeomorphic flow is constructed as a compressible flow based on the volume forms at the manifold. Such a flow yields equality of each local volume element between the original manifold and the target at its final state. Furthermore, the salient features can be used to register the manifold to a reference template by an incompressible flow guided by a divergence-free vector field within the manifold. The process can retain the equality of local volume elements while registering the manifold to a template at the same time. An efficient and practical algorithm is also presented to generate a volume-preserving mapping and a salient feature registration on discrete 3D volumes which are represented with tetrahedral meshes embedded in 3D space. This method can be applied to comparative analysis and visualization of volumetric medical imaging data across subjects. We demonstrate an example application in multimodal neuroimaging data analysis and collective data visualization.

Parameterization of complex surfaces constitutes a major means of visualizing highly convoluted geometric structures as well as other properties associated with the surface. It also enables users with the ability to navigate, orient, and focus on regions of interest within a global view and overcome the occlusions to inner concavities. In this paper, we propose a novel area-preserving surface parameterization method which is rigorous in theory, moderate in computation, yet easily extendable to surfaces of non-disc and closed-boundary topologies. Starting from the distortion induced by an initial parameterization, an area restoring diffeomorphic flow is constructed as a Lie advection of differential 2-forms along the manifold, which yields equality of the area elements between the domain and the original surface at its final state. Existence and uniqueness of result are assured through an analytical derivation. Based upon a triangulated surface representation, we also present an efficient algorithm in line with discrete differential modeling. As an exemplar application, the utilization of this method for the effective visualization of brain cortical imaging modalities is presented. Compared with conformal methods, our method can reveal more subtle surface patterns in a quantitative manner. It, therefore, provides a competitive alternative to the existing parameterization techniques for better surface-based analysis in various scenarios.

This paper formalizes a novel, intrinsic geometric scale space (IGSS) of 3D surface shapes. The intrinsic geometry of a surface is diffused by means of the Ricci flow for the generation of a geometric scale space. We rigorously prove that this multiscale shape representation satisfies the axiomatic causality property. Within the theoretical framework, we further present a feature-based shape representation derived from IGSS processing, which is shown to be theoretically plausible and practically effective. By integrating the concept of scale-dependent saliency into the shape description, this representation is not only highly descriptive of the local structures, but also exhibits several desired characteristics of global shape representations, such as being compact, robust to noise and computationally efficient. We demonstrate the capabilities of our approach through salient geometric feature detection and highly discriminative matching of 3D scans.

Abstract This paper develops a new salient keypoints‐based shape description which extracts the salient surface keypoints with detected scales. Salient geometric features can then be defined collectively on all the detected scale normalized local patches to form a shape descriptor for surface matching purpose. The saliency‐driven keypoints are computed as local extrema of the difference of Gaussian function defined over a curved surface in geodesic scale space. This method can properly function on either manifold or non‐manifold surface without resorting to any surface mapping or parameterization procedures. Therefore, it has a wide utility in many applications such as shape matching, classification, and recognition. Our experiments on 3D shapes demonstrate that the salient keypoints and local feature descriptors are robust and stable to noisy input and insensitive to resolution change. We have applied our technique to the tasks of 3D shape matching, and the experimental results showed good performance and the effectiveness of this new method. Copyright © 2008 John Wiley & Sons, Ltd.

This paper presents a novel integrative information visualization framework for cross-subject neuroimaging data analysis. The framework can integrate multimodal information captured by different imaging modalities and population-based statistical information presented by different subjects. In this framework, accurate registration of cortical structures is the foundation for the information integration across population. We present a non-rigid intersubject brain surface registration method using conformal structure and spherical thin-plate splines. Spherical thin-plate splines are designed to explicitly match prominent homologous landmarks, and meanwhile, interpolate a global deformation field on the spherical domain, registering brain surfaces in a transformed space. Subsequently, an approach for the integrative information fusion and visualization is presented to handle multimodality neuroimaging data. The entire framework demonstrates its usefulness in multimodality neuroimaging data analysis across subjects.

Recent advances in imaging technologies, such as magnetic resonance imaging (MRI), positron emission tomography (PET) and diffusion tensor imaging (DTI) have accelerated brain research in many aspects. In order to better understand the synergy of the many processes involved in normal brain function, integrated modeling and analysis of MRI, PET, and DTI across subjects is highly desirable. The current state-of-art computational tools fall short in offering an analytic approach for intersubject brain registration and analysis. In this paper we present an approach which is based on landmark constrained conformal parameterization of a brain surface from high-resolution structural MRI data to a canonical spherical domain. This model allows natural integration of information from co-registered PET as well as DTI data and lays a foundation for the quantitative analysis of the relationship among diverse datasets across subjects. Consequently, the approach can be extended to provide a software environment able to facilitate detection of abnormal functional brain patterns in patients with neurological disorder.