Knowledge-Based Systems最新文献

Designing effective coverage routes for unmanned surface vehicles (USVs) is crucial to improve the efficiency of offshore bathymetric surveys. However, existing coverage planning methods for practical use are limited, primarily due to the large-scale surveying areas and intricate region geometries caused by coastal features. This study aims to address these challenges by introducing a coverage path planning framework for USV-assisted bathymetric mapping, specifically aimed at the joint optimization of paths to cover numerous complex regions. Initially, we conceptualize the large-scale bathymetric survey mission as an integer programming model. The model uses four distinct decision variables to meticulously formulate length calculations, inter-regional connections, entry and exit point selections, and line sweep direction. Then, a novel hierarchical algorithm is devised to solve the problem. The method first incorporates a bisection-based convex decomposition method to achieve optimal partitioning of complex regions. Additionally, a hierarchical heuristic optimization algorithm that seamlessly integrates the optimization of all influencing factors is designed, which includes order generation, candidate pattern finding, tour finding, and final optimization. The reliability of the framework is validated through semi-physical simulations and lake trials using a real USV. Through comparative studies, our model demonstrates clear advantages in computational efficiency and optimization capability compared to state-of-the-arts, with its superiority becoming more pronounced as the problem scale increases. The results from lake trials further affirm the efficient and reliable performance of our model in practical bathymetric survey tasks.

A taxonomy represents a hierarchically structured knowledge graph that forms the infrastructure for various downstream applications, including recommender systems, web search, and question answering. The exploration of automated induction from text corpora has yielded notable taxonomies such as CN-probase, CN-DBpedia, and Zhishi.schema. Despite these efforts, existing taxonomies still face two critical issues that result in sub-optimal hierarchical structures. On the one hand, commonly observed taxonomies exhibit a coarse-grained and “flat” structure, stemming from a noticeable lack of diversity in both nodes and edges. This limitation primarily originates from the biased and homogeneous data distribution. On the other hand, the semantic granularity among “siblings” within these taxonomies remains inconsistent, presenting a challenge in accurately and comprehensively identifying hierarchical relations. To address these issues, this study introduces a novel taxonomy induction framework composed of three meticulously designed components. Initially, we established a seed schema by leveraging statistical information from external data sources as distant supervision to append nodes and edges containing “generic semantics”, thereby rectifying biased data distributions. Subsequently, a clustering algorithm is employed to group the nodes based on their similarities, followed by a refinement operation of the hierarchical relations among these nodes. Building on this seed schema, we propose a fine-grained contrastive learning method in the expansion module to strengthen the utilization of taxonomic structures, consequently boosting the precision of query-anchor matching. Finally, we meticulously scrutinized the hierarchical relations between each query and its siblings to ensure the integrity of the constructed taxonomy. Extensive experiments on real-world datasets validated the efficacy of our proposed framework for constructing well-structured taxonomies.

Recently, there has been an increasing interest in extracting and annotating tables on the Web. This activity allows the transformation of textual data into machine-readable formats to enable the execution of various artificial intelligence tasks, e.g., semantic search and dataset extension. Semantic Table Interpretation (STI) is the process of annotating elements in a table. The paper explores Semantic Table Interpretation, addressing the challenges of Entity Retrieval and Entity Disambiguation in the context of Knowledge Graphs (KGs). It introduces LamAPI, an Information Retrieval system with string/type-based filtering and s-elBat, an Entity Disambiguation technique that combines heuristic and ML-based approaches. By applying the acquired know-how in the field and extracting algorithms, techniques and components from our previous STI approaches and the state of the art, we have created a new platform capable of annotating any tabular data, ensuring a high level of quality.

Swarm learning is an emerging technique for collaborative machine learning in which several participants train machine learning models without sharing private data. In a standard swarm network, all the nodes in the network receive identical final models regardless of their individual contributions. This mechanism may be deemed unfair from an economic perspective, discouraging organizations with more resources from participating in any collaboration. Here, we present a framework for swarm learning in which nodes receive personalized models based on their contributions. The results of this study demonstrate the efficacy of this approach by showing that all participants experience performance enhancements compared to their local models. However, participants with higher contributions receive better models than those with lower contributions. This fair mechanism results in the highest possible accuracy for the most contributive participant, comparable to the standard swarm learning model. Such incentive structure can motivate resource-rich organizations to engage in collaboration, leading to the development of machine learning models that incorporate data from more resources, which is ultimately beneficial for every party.

Modeling actor interaction relations is crucial for group activity recognition. Previous approaches often adopt a fixed paradigm that involves calculating an affinity matrix to model these interaction relations, yielding significant performance. On the one hand, the affinity matrix introduces an inductive bias that actor interaction relations should be dynamically computed based on the input actor features. On the other hand, MLPs with static parameterization, in which parameters are fixed after training, can represent arbitrary functions. Therefore, it is an open question whether inductive bias is necessary for modeling actor interaction relations. To explore the impact of this inductive bias, we propose an affinity matrix-free paradigm that directly uses the MLP with static parameterization to model actor interaction relations. We term this approach MLP-AIR. This paradigm overcomes the limitations of the inductive bias and enhances the capture of implicit actor interaction relations. Specifically, MLP-AIR consists of two sub-modules: the MLP-based Interaction relation modeling module (MLP-I) and the MLP-based Relation refining module (MLP-R). MLP-I is used to model the spatial–temporal interaction relations by emphasizing cross-actor and cross-frame feature learning. Meanwhile, MLP-R is used to refine the relation between different channels of each relation feature, thereby enhancing the expression ability of the features. MLP-AIR is a plug-and-play module. To evaluate our module, we applied MLP-AIR to replicate three representative methods. We conducted extensive experiments on two widely used benchmarks—the Volleyball and Collective Activity datasets. The experiments demonstrate that MLP-AIR achieves favorable results. The code is available at https://github.com/Xuguoliang12/MLP-AIR.

To inhibit the spread of rumorous information and its severe impacts, fact checking aims at retrieving relevant evidence to verify the veracity of a given statement. Fact checking methods typically use knowledge graphs (KGs) as external repositories and develop reasoning mechanism to retrieve evidence for verifying the statement. Existing fact checking methods have focused on verifying the statement of a single claim expressed by a clause. However, as real-world rumorous information is usually complex and a textual statement is often composed of multiple clauses (i.e. represented as multiple claims instead of a single one), multi-claim fact checking is not only necessary but more important for practical applications. Multi-claim statements imply rich contextual information and modeling the interactions of multiple claims can facilitate better verification. In this paper, we propose a knowledge enhanced learning and semantic composition model for multi-claim fact checking. Our model consists of two modules, KG-based learning enhancement and multi-claim semantic composition. To fully utilize the contextual information implied in multiple claims, the KG-based learning enhancement module learns the dynamic context-specific representations via selectively aggregating relevant attributes of entities. To robustly verify multiple claims robustly, the multi-claim semantic composition module learns a unified representation for multiple claims by modeling inter-claim interactions, and then verify them as a whole on the basis of this. We conduct experimental studies to validate our proposed method, and the experimental results on three typically datasets confirmed the efficacy of our model for multi-claim fact checking.

Continuous-Time Dynamic Graph (CTDG) methods have shown their superior ability in learning representations for dynamic graph-structured data, the methods split the sequential updating process into discrete batches to reduce the computation costs, as a result, the message constructor in existing CTDG methods cannot be optimized by gradient descent and is designed to be parameter-free. In particular, this layer fails to embed complex event subgraphs and ignores the structure information, while most real-world events are structured and complex. For example, a paper publication event in an academic graph contains different relations like authorship and citations. Furthermore, the corresponding nodes could not receive position-wise messages to make precise representation updates. To tackle this issue, we propose a new method called Temporal Graph Network for continuous-time dynamic Event sequence (TGNE) with a structure-aware message constructor to update node representation with complex event subgraph, by treating message construction and delivery as a message-passing process, in this way, the message constructor can be formalized as a graph neural network layer. TGNE extends the input of CTDG methods to subgraphs with complex structures and preserves more information in message delivery. Extensive experiments demonstrate that the proposed method can achieve competitive performance on traditional tasks on bipartite graphs and event sequence learning tasks on heterogeneous graphs.

Most methods only use normal samples to learn anomaly detection (AD) models in an unsupervised manner. However, these samples may be noisy in real-world applications, causing the models to be unable to accurately identify anomaly objects. In addition, there are a small number of anomaly samples in real industrial production that should be fully utilized to help model discrimination. Existing methods of introducing anomaly samples still have bottlenecks in model identification capabilities. In this paper, by introducing both normal and a few abnormal samples, we propose a novel semi-supervised learning method for anomaly detection, named RobustPatch, which can improve the model discriminability through a self-cross scoring mechanism and the learning of feature AutoEncoder. Our approach contains two core designs: Firstly, we propose a self-cross scoring module, calculating the weights of normal and anomaly features extracted from corresponding images using a self-scoring and cross-scoring manner, respectively. Secondly, our approach proposes a fully connected feature AutoEncoder to rate the extracted features, which is trained with the supervision of the scored weights. Extensive experiments on the MVTecAD and BTAD datasets validate the superior anomaly boundaries discriminability of our approach and superior performance in noise-polluted scenarios.

Recently, remote sensing images have become popular in various tasks, including resource exploration. However, limited by hardware conditions and formation processes, the obtained remote sensing images often suffer from low-resolution problems. Unlike the high cost of hardware to acquire high-resolution images, super-resolution software methods are good alternatives for restoring low-resolution images. In addition, remote sensing images have a common nature that similar visual patterns repeatedly appear across distant locations. To fully capture these long-range satellite image contexts, we first introduce the global attention network super-resolution method to reconstruct the images. This network improves the performance but introduces unessential information while significantly increasing the computational effort. To address these problems, we propose an innovative method named the global sparse attention network (GSAN) that integrates both sparsity constraints and global attention. Specifically, our method applies spherical locality sensitive hashing (SLSH) to convert feature elements into hash codes, constructs attention groups based on the hash codes, and computes the attention matrix according to similar elements in the attention group. Our method captures valid and useful global information and reduces the computational effort from quadratic to asymptotically linear regarding the spatial size. Extensive qualitative and quantitative experiments demonstrate that our GSAN has significant competitive advantages in terms of performance and computational cost compared with other state-of-the-art methods.

Backdoor attacks in the traditional graph neural networks (GNNs) field are easily detectable due to the dilemma of confusing labels. To explore the backdoor vulnerability of GNNs and create a more stealthy backdoor attack method, a clean-label graph backdoor attack method(CGBA) in the node classification task is proposed in this paper. Differently from existing backdoor attack methods, CGBA requires neither modification of node labels nor graph structure. Specifically, to solve the problem of inconsistency between the contents and labels of the samples, CGBA selects poisoning samples in a specific target class and uses the samples’ own label as the target label (i.e., clean-label) after injecting triggers into the target samples. To guarantee the similarity of neighboring nodes, the raw features of the nodes are elaborately picked as triggers to further improve the concealment of the triggers. Extensive experiments results show the effectiveness of our method. When the poisoning rate is 0.04, CGBA can achieve an average attack success rate of 87.8%, 98.9%, 89.1%, and 98.5%, respectively.