2026-05-30 2026, Volume 2 Issue 1

  • Select all
  • research-article
    Guanghuang Liu
    2026, 2 (1): 1-20 . https://doi.org/10.71204/hefd5f67

    As poverty reduction strategies shift comprehensively toward alleviating relative poverty, precisely identifying multidimensional relative poverty populations has become critical for social governance. However, existing data-driven models often face algorithmic bottlenecks—such as spatial heterogeneity, regional sample sparsity, and extreme category imbalance—when applied to complex scenarios characterized by vast territories and significant regional disparities. To address this, this study proposes a two-stage knowledge-distilled adaptive gradient boosting framework (TKDAF). First, in the prior knowledge extraction stage (Stage I), a base structure-regularized gradient boosting tree model is constructed. Combined with SHAP game-theoretic attribution, this stage quantifies and extracts the global objective weights of poverty-inducing features across regions. Second, in the spatial adaptive enhancement stage (Stage II), the S-DAGB (Spatial-Adaptive Distilled Gradient Boosting) core prediction model is introduced. It achieves deep integration of multiple regularization and feature enhancement mechanisms by incorporating: feature space nonlinear reconstruction based on prior knowledge, a dynamic category weighting mechanism based on effective sample size (ENS), and spatial adaptive optimization using the TPE Bayesian algorithm.Empirical results based on the 2020 China Family Panel Studies (CFPS) multidimensional dataset demonstrate that the S-DAGB model not only effectively overcomes the generalization bottleneck of deep tree models in the sample-constrained Northeast region (achieving 93.52% accuracy),but also significantly improves precision in regions with highly heterogeneous features and extreme class imbalance, such as central and western China. This effectively reduces wasteful allocation of poverty alleviation resources caused by false positive errors. This study provides an algorithmic solution that combines high accuracy with interpretability for precise identification of relative poverty in complex data distribution scenarios.

  • research-article
    Yingchen Xu, Ye Tian, Ying Jing, Fujiang Yuan
    2026, 2 (1): 21-36 . https://doi.org/10.71204/pyppdf95

    The Practical Byzantine Fault-Tolerant (PBFT) consensus algorithm faces challenges such as random master node selection, high communication overhead, and a lack of incentive and penalty mechanisms, which undermine its efficiency and security in large-scale network environments. To address these issues, this paper proposes an improved PBFT consensus algorithm, RF-PBFT, which integrates a random forest-based node grouping mechanism with a dynamic reputation system. The algorithm leverages key behavioral data of nodes during the consensus process to train a random forest model, partitioning nodes into a consensus set and a candidate set. Nodes with superior predictive performance are selected to participate in the consensus process, thereby reducing redundant communication overhead. Furthermore, a differentiated reputation mechanism is established for the consensus group and the candidate group to encourage positive behavior and deter malicious actions. The master node is elected through a combination of prediction results and voting, enhancing system reliability and security. Simulation results demonstrate that, compared with traditional PBFT and other state-of-the-art variants, RF-PBFT significantly reduces communication overhead, decreases consensus latency, and improves throughput, validating its effectiveness in multi-node blockchain systems.

  • research-article
    Hongxia Su
    2026, 2 (1): 37-50 . https://doi.org/10.71204/ew979106

    To address the challenges of vehicle detection accuracy in nighttime UAV aerial photography scenarios caused by low light conditions, strong noise, and dense small object distribution, this paper proposes an improved YOLO11 vehicle detection method based on multi-modal fusion. First, a collaborative enhancement strategy combining CLAHE and Gamma correction is applied to preprocessing nighttime visible light images, effectively restoring vehicle texture details in dark areas. Second, the enhanced visible light images are fused with infrared images through fixed-weight weighted fusion (infrared weight α=0.7), fully leveraging the complementary advantages of both modalities. Finally, the MANet module is introduced on top of YOLO11n to enhance backbone multi-scale feature extraction capabilities, while the ADFPN-DASI module is designed to collaboratively optimize neck multi-scale feature representation. Experiments conducted on the DroneVehicle dataset demonstrate that the proposed method achieves an mAP50 of 80.53%, representing an improvement of 11.45 percentage points over the YOLO11n baseline. The method maintains lightweight advantages in parameter and computational resources, outperforming mainstream comparison methods such as YOLOv5n.

  • research-article
    Shufeng Wang
    2026, 2 (1): 51-70 . https://doi.org/10.71204/vwpczb65

    With the rapid advancement and widespread deployment of artificial intelligence, issues related to ethics, trust, and security have become increasingly prominent, posing significant challenges to its sustainable development. This paper aims to systematically investigate these intertwined challenges by constructing a structured analytical framework encompassing three core dimensions: ethics, trust, and security.Methodologically, the study analyzes the evolutionary trajectory and application contexts of AI, and synthesizes key risk patterns and governance concerns. It examines major ethical issues, including algorithmic bias, data privacy, responsibility attribution, and value alignment, and further explores the underlying tension between technological rationality and social values.In addition, the paper proposes a multi-level trust formation mechanism based on technological reliability, institutional assurance, and public cognition, emphasizing that trust emerges from the joint interaction of technical systems and governance structures. At the security level, it highlights the complexity and cross-domain nature of AI risks, and advocates for a full life-cycle governance framework integrating risk assessment, technical auditing, and continuous supervision.The study’s key contribution lies in integrating ethical, trust, and security perspectives into a unified analytical framework, and in proposing a collaborative governance approach involving governments, enterprises, and international stakeholders.

  • research-article
    Chaohui Zhang, Lujie Ren, Jianming Mao, Haijun Lei, Dengcheng Lu, Shishui Zhou
    2026, 2 (1): 71-79 . https://doi.org/10.71204/fef1s997

    Robot joint encoders are critical sensing components in industrial robots, where positioning accuracy, dynamic response, and environmental adaptability directly affect motion-control performance and operational reliability. However, conventional encoder technologies still face challenges related to signal instability under harsh industrial conditions, insufficient high-speed response capability, and limited long-term reliability. To address these issues, this paper presents the design and implementation of a non-contact magnetic sensing and signal-processing system for high-precision robot joint encoders. The proposed system integrates incremental magnetic grid design, tunnel magnetoresistance (TMR) sensing technology, adaptive signal conditioning, multi-dimensional error compensation, and high-speed signal-processing algorithms into a unified encoder framework. A dual-reading-head magnetic grid structure and dynamic gain compensation mechanism are introduced to improve signal stability under vibration, temperature variation, oil contamination, and electromagnetic interference conditions. In addition, adaptive filtering and real-time interpolation algorithms are employed to enhance dynamic response capability and suppress nonlinear measurement errors during high-speed robot motion. The developed encoder system supports multiple industrial communication interfaces, including RS422, SSI, and CANopen, enabling flexible integration with industrial robot control systems. Experimental results demonstrate that the proposed encoder achieves repeatability accuracy within ±1 μm and a frequency response of 8000 kHz while maintaining stable operation under harsh industrial environments. Engineering deployment in industrial robots and precision manufacturing equipment further verifies the robustness, reliability, and practical applicability of the proposed system. The research provides an effective technical solution for high-precision robotic motion sensing and intelligent industrial measurement systems.

Publishing model
0

Downloads

Submission to first decision (median)
10 days

ISSN 3068-1448 (Online)