Identifying anomalous trajectories that deviate from usual driving patterns in an open-world context has recently become a critical and urgent task in location-aware systems. In contrast to the closed-world settings, ...
详细信息
Identifying anomalous trajectories that deviate from usual driving patterns in an open-world context has recently become a critical and urgent task in location-aware systems. In contrast to the closed-world settings, its goal is to not only identify known but also detect the presence of unknown anomalous behaviors. Despite the achievements of recent self-motivated learning paradigms, existing solutions either do not have the ability to discover unknown anomalous behaviors or can only expose their presence. In this study, we first formally define our task as the Anomalous Trajectory Discovery problem in an Open-world scenario (ATDO), and introduce a novel fine-grained O pen-world S tate S pace L earning ( OSSL ) framework that has the ability to discover multiple unknown patterns in addition to classifying the existing known behavioral patterns of trajectories. Due to the inherent density behind massive trajectories, we are motivated by state space models and devise a Spatial- temporal State Space (S3) block to explore the long-term dependencies behind the lengthy trajectories. To enable open-world learning, we devise two adapters that operate interactively with three open-world objectives to discover multiple unknown patterns in addition to identifying existing behaviors. With the context setting of two progressive open-world tasks, the experimental results conducted on two large trajectory datasets demonstrate the superiority of OSSL for both known and unknown anomalous patterns.
This paper presents BC-SBOM, a novel blockchainbased system designed to enhance the management of software Bills of Materials (SBOMs). By leveraging blockchain technology, BC-SBOM ensures secure storage and sharing of...
详细信息
ISBN:
(纸本)9791188428137
This paper presents BC-SBOM, a novel blockchainbased system designed to enhance the management of software Bills of Materials (SBOMs). By leveraging blockchain technology, BC-SBOM ensures secure storage and sharing of SBOMs, while providing a comprehensive global view of dependencies among software components. The system also supports rapid propagation of alerts for newly discovered vulnerabilities, thereby increasing responsiveness to potential threats. Offering superior reliability, transparency, and availability compared to traditional SBOM tools, BC-SBOM aims to significantly improve the management of complex software systems and contribute to the advancement of software security practices. Copyright 2025 Global IT Research Institute (GIRI). All rights reserved.
Existing Automated Service Composition (ASC) approaches typically require inputs to be in a designated form. These, namely tuples, pose challenges due to the significant divergence from the most commonly used and stra...
详细信息
Implementing new features in repository-level codebases is a crucial application of code generation models. However, current benchmarks lack a dedicated evaluation framework for this capability. To fill this gap, we i...
详细信息
Missing data is a prevalent challenge in real-world applications, hindering the usability and quality of datasets. Data imputation, a method to substitute missing values, offers a solution. While existing techniques a...
详细信息
Missing data is a prevalent challenge in real-world applications, hindering the usability and quality of datasets. Data imputation, a method to substitute missing values, offers a solution. While existing techniques achieve promising results, they often rely on centralized data collection, raising privacy concerns. Existing methods for data imputation mainly utilize heterogeneous federation or optimization methods such as meta-learning. Although they can achieve personalized imputation, since they do not consider the impact of different data missing rates on the imputation quality, the imputed data may perform poorly in subsequent tasks. To address this, we propose a personalized federated approach FedImpute, from the perspective of data level. FedImpute tackles the challenge of balancing global model performance with local data customization to achieve robust imputation, whose core concept lies in leveraging the strengths of both global and local data perspectives. The global model captures universal patterns, while local models adapt to the unique characteristics of each participant's private data. To achieve this, FedImpute incorporates modules, including the clusterer and auxiliary classifier, to extract and utilize latent class information during the imputation process. This enables the model to prioritize similarities within similar categories, leading to more precise and personalized imputations. Extensive evaluations on four real-world datasets demonstrate that FedImpute overall outperforms existing methods, especially for high missing situations.
[Context & Motivation] Explainable autonomous systems are increasingly essential for engendering trust, especially when they are deployed in safety-critical scenarios. [Question/Problem] Despite the robust reliabi...
详细信息
In this correspondence, a reconfigurable intelligent surface (RIS) assisted wireless power transfer system where a multi-antenna energy station (ES) transmits energy beams to multi-antenna energy receivers (ERs) is co...
详细信息
In this correspondence, a reconfigurable intelligent surface (RIS) assisted wireless power transfer system where a multi-antenna energy station (ES) transmits energy beams to multi-antenna energy receivers (ERs) is considered. We aim at optimizing the ES beamforming and the RIS beamforming to minimize the transmit power of the ES under individual energy harvesting constraints for ERs. To handle the beamforming optimization problem, we develop an algorithm where an alternating direction method of multipliers based scheme is put forward for the RIS beamforming optimization. Compared with the counterpart which relies on semidefinite relaxation to optimize the RIS beamforming, the complexity of the proposed beamforming optimization algorithm is significantly reduced. Simulation results illustrate that the proposed algorithm performs close to the counterpart, which implies that the complexity reduction is achieved at very limited expense of performance.
Space deployable antenna is the key equipment in realizing the communication and data transmission between the spacecraft and the *** order to enrich the configurations of deployable antennas,the type synthesis of dep...
详细信息
Space deployable antenna is the key equipment in realizing the communication and data transmission between the spacecraft and the *** order to enrich the configurations of deployable antennas,the type synthesis of deployable mechanisms for ring truss antenna is conducted in this ***,the principle of the constraint-synthesis method based on screw theory is briefly described,the structure of the ring truss deployable antenna and its folding principle are analyzed,and the ring truss mechanism is divided into upper edges,lower edges and ***,based on the constraint-synthesis method,the type synthesis of the basic unit edges is carried out,a series of basic unit mechanisms are obtained from combining the basic unit edge mechanisms,and five mechanism units with fewer joints and simple structures are ***,simulation models of the five ring truss deployable mechanisms are built in Solidworks and Matlab software,and the deploying process is verified by the movement ***,mechanism characteristics of the five mechanisms are analyzed and discussed,and a prototype is manufactured,verifying the analysis in this *** research provides a new way for the type synthesis of spatial deployable mechanisms,and the ring truss deployable mechanisms obtained in this study can be well applied in the field of aerospace.
The reliability of large language models remains a critical challenge, particularly due to their susceptibility to hallucinations and factual inaccuracies during text generation. Existing solutions either underutilize...
详细信息
Today, GPUs significantly boost rendering performance. However, the high memory requirements limit their use, especially on low-end mobile platforms. Compression techniques have been widely adopted to reduce memory co...
详细信息
Today, GPUs significantly boost rendering performance. However, the high memory requirements limit their use, especially on low-end mobile platforms. Compression techniques have been widely adopted to reduce memory consumption but face two primary issues when applied to mobile GPUs: (1) low repetition ratio caused by small raw data sizes and concurrency, and (2) low locality caused by unpredictable rendering behaviors. These two limitations result in a low compression ratio when compressors are applied to low-end mobile devices. This article introduces gCom, a fine-grained rendering compressor accelerated by GPUs. To improve the compression ratio, gCom incorporates the following innovations. First, unlike other compression techniques that use frames or tiles as basic processing units, gCom is the first to employ a fine-grained processing unit (i.e., the color channel), enhancing repetition amplification without increasing raw data. Second, gCom introduces two key features-Hierarchical Delta and Channel Decorrelator-which maximize the locality of adjacent channels and reduce raw data size. Third, to maintain the original GPU throughput, gCom revolutionizes the Golomb-Rice algorithm and proposes a new compression approach, the parallel-Oriented Golomb-Rice algorithm, enabling parallel execution of both decompression and compression processes. The entire design of gCom utilizes only idle resources and existing commands on mobile GPUs, thus keeping purchasing costs low. To date, gCom has improved the channel locality by nearly 50%. The best compression achievement received by gCom has reached around 20%.
暂无评论