检索结果-内蒙古大学图书馆

A deep embedded clustering technique using dip test and unique neighbourhood set

Neural computing and Applications 2025年第3期37卷 1345-1356页

作者： Rahman, Md Anisur Ang, Li-Minn Sun, Yuan Seng, Kah Phooi La Trobe Business School La Trobe University Plenty Road BundooraVIC3086 Australia School of Science Technology and Engineering University of the Sunshine Coast Sippy DownsQLD4556 Australia School of AI & Advanced Computing Xi’an Jiaotong Liverpool University Suzhou215123 China

In recent years, there has been a growing interest in deep learning-based clustering. A recently introduced technique called DipDECK has shown effective performance on large and high-dimensional datasets. DipDECK utilises Hartigan’s dip test, a statistical test, to merge small non-viable clusters. Notably, DipDECK was the first deep learning-based clustering technique to incorporate the dip test. However, the number of initial clusters of DipDECK is overestimated and the algorithm then randomly selects the initial seeds to produce the final clusters for a dataset. Therefore, in this paper, we presented a technique called UNSDipDECK, which is an improved version of DipDECK and does not require user input for datasets with an unknown number of clusters. UNSDipDECK produces high-quality initial seeds and the initial number of clusters through a deterministic process. UNSDipDECK uses the unique closest neighbourhood and unique neighbourhood set approaches to determine high-quality initial seeds for a dataset. In our study, we compared the performance of UNSDipDECK with fifteen baseline clustering techniques, including DipDECK, using NMI and ARI metrics. The experimental results indicate that UNSDipDECK outperforms the baseline techniques, including DipDECK. Additionally, we demonstrated that the initial seed selection process significantly contributes to UNSDipDECK ’s ability to produce high-quality clusters. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Explore Statistical Properties of Undirected Unweighted Networks from Ensemble Models 27th

Explore Statistical Properties of Undirected Unweighted Netw...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Zhao, Xunda Wu, Xing Wang, Jianjia School of Computer Engineering and Science Shanghai University Shanghai China Engineering and Science Shanghai Institute for Advanced Communication and Data Science School of Computer Shanghai University Shanghai China School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou China

ISBN: (纸本)9783031783975

Complex network theory has been widely demonstrated as a powerful tool in modeling and characterizing various complex systems. In the past, complex network theory has focused on the behaviors as well as the characteristics of the network nodes and edges. However, with the continuous evolution of society, traditional graph theory faces challenges due to the emergence of extermely large network structures. Recently, complex network method based statistics has attracted much attention. The new approach effectively manages very large networks and uncovers their intrinsic properties. In this paper, we present a complex network analysis model for undirected, unweighted networks based on a statistical analysis approach. This model is inspired by the ensemble model in thermostatistical physics. Based on the established mathematical model, we derive physical measures that reflect the intrinsic properties of the network, including Entropy, Free Energy, Temperature, and so on. In the experimental part, we first explored the mathematical characterization of these metrics. Then, we observed the performance of various network categories under the same metric. Finally, we applied these measures to the field of graph classification. Extensive experiments demonstrate the effectiveness and superiority of the proposed method. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Undirected graphs

来源：评论

学校读者我要写书评

暂无评论

EFFICIENT TRIANGULAR ARBITRAGE DETECTION VIA GRAPH NEURAL NETWORKS

arXiv

引用

arXiv 2025年

作者： Zhang, Di School of AI And Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou215123 China

Triangular arbitrage is a profitable trading strategy in financial markets that exploits discrepancies in currency exchange rates. Traditional methods for detecting triangular arbitrage opportunities, such as exhaustive search algorithms and linear programming solvers, often suffer from high computational complexity and may miss potential opportunities in dynamic markets. In this paper, we propose a novel approach to triangular arbitrage detection using Graph Neural Networks (GNNs). By representing the currency exchange network as a graph, we leverage the powerful representation and learning capabilities of GNNs to identify profitable arbitrage opportunities more efficiently. Specifically, we formulate the triangular arbitrage problem as a graph-based optimization task and design a GNN architecture that captures the complex relationships between currencies and exchange rates. We introduce a relaxed loss function to enable more flexible learning and integrate Deep Q-Learning principles to optimize the expected returns. Our experiments on a synthetic dataset demonstrate that the proposed GNN-based method achieves a higher average yield with significantly reduced computational time compared to traditional methods. This work highlights the potential of using GNNs for solving optimization problems in finance and provides a promising approach for real-time arbitrage detection in dynamic financial markets. © 2025, CC BY-NC-SA.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

DaCoM: Strategies to Construct Domain-specific Low-resource Language Machine Translation Dataset 31

DaCoM: Strategies to Construct Domain-specific Low-resource ...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Kang, Junghoon Tak, Keunjoo Choi, Joung Su Kim, Myunghyun Jang, Junyoung Kang, Youjin AI Center HD Korea Shipbuilding & Offshore Engineering Korea Republic of DT Innovation Deparment HD Hyundai Samho Korea Republic of School of Computing Korea Advanced Institute of Science & Technology Korea Republic of

ISBN: (纸本)9798891761971

Translation of low-resource languages in industrial domains is essential for improving market productivity and ensuring foreign workers have better access to information. However, existing translators struggle with domain-specific terms, and there is a lack of expert annotators for dataset creation. In this work, we propose DaCoM, a methodology for collecting low-resource language pairs from industrial domains to address these challenges. DaCoM is a hybrid translation framework enabling effective data collection. The framework consists of a large language model and neural machine translation. Evaluation verifies existing models perform inadequately on DaCoM-created datasets, with up to 53.7 BLEURT points difference depending on domain inclusion. DaCoM is expected to address the lack of datasets for domain-specific low-resource languages by being easily pluggable into future state-of-the-art models and maintaining an industrial domain-agnostic approach. ©2025 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Machine Learning for Predictive Risk Analytics in Industry 4.0: A Comprehensive Random Forest Evaluation 2nd

Machine Learning for Predictive Risk Analytics in Industry 4...

引用

2nd International Conference on Intelligent Manufacturing and Robotics, ICIMR 2024

作者： Asante, Oliver J. Rusliem, Enrico W. Pan, Yue Basarah, Michelle A. Alin, Cashmere B. R. Zhang, Guangfeng Chen, Yi School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou China Zhuhai China School of Intelligent Manufacturing Ecosystem Xi’an Jiaotong-Liverpool University Suzhou China

ISBN: (纸本)9789819639489

In financial forecasting, where navigating uncertainties is crucial for informed decision-making, risk management plays an essential role. Large datasets and market fluctuations can lead to inaccurate forecasts and significant losses. Fortunately, artificial intelligence and machine learning algorithms offer promising solutions. The Random Forest Model, a commonly used machine learning algorithm, has the potential to alleviate these issues. Unlike conventional models that rely on a single algorithm, Random Forest aggregates multiple decision trees, each analysing a subset of data and identifying key variables. Averaging the predictions using this method appeared to reduce bias, improve accuracy, and make it adept at handling messy financial data. This paper aims to provide an in-depth analysis of Random Forest’s potential in financial forecasting and risk management, especially evaluating its effectiveness and practicality in the field. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Lossless Acceleration of Large Language Models with Hierarchical Drafting based on Temporal Locality in Speculative Decoding

arXiv

引用

arXiv 2025年

作者： Cho, Sukmin Choi, Sangjin Hwang, Taeho Seo, Jeongyeon Jeong, Soyeong Lee, Huije Song, Hoyun Park, Jong C. Kwon, Youngjin School of Computing Korea Advanced Institute of Science and Technology Korea Republic of Graduate School of AI Korea Advanced Institute of Science and Technology Korea Republic of

Accelerating inference in Large Language Models (LLMs) is critical for real-time interactions, as LLMs have been widely incorporated into real-world services. Speculative decoding, a fully algorithmic solution, has gained attention for improving inference speed by drafting and verifying tokens, thereby generating multiple tokens in a single forward pass. However, current drafting strategies usually require significant fine-tuning or have inconsistent performance across tasks. To address these challenges, we propose Hierarchy Drafting (HD)1, a novel lossless drafting approach that organizes various token sources into multiple databases in a hierarchical framework based on temporal locality. In the drafting step, HD sequentially accesses multiple databases to obtain draft tokens from the highest to the lowest locality, ensuring consistent acceleration across diverse tasks and minimizing drafting latency. Our experiments on Spec-Bench using LLMs with 7B and 13B parameters demonstrate that HD outperforms existing lossless drafting methods, achieving robust inference speedups across model sizes, tasks, and temperatures. © 2025, CC BY-NC-SA.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Learnable Prompting SAM-induced Knowledge Distillation for Semi-supervised Medical Image Segmentation

引用

IEEE Transactions on Medical Imaging 2025年第5期44卷 2295-2306页

作者： Huang, Kaiwen Zhou, Tao Fu, Huazhu Zhang, Yizhe Zhou, Yi Gong, Chen Liang, Dong Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing210094 China Agency for Science Technology and Research Institute of High-Performance Computing Singapore Singapore Southeast University School of Computer Science and Engineering Nanjing211189 China Chinese Academy of Sciences Lauterbur Research Center for Biomedical Imaging and the Research Center for Medical AI Shenzhen Institute of Advanced Technology Shenzhen518055 China

The limited availability of labeled data has driven advancements in semi-supervised learning for medical image segmentation. Modern large-scale models tailored for general segmentation, such as the Segment Anything Model (SAM), have revealed robust generalization capabilities. However, applying these models directly to medical image segmentation still exposes performance degradation. In this paper, we propose a learnable prompting SAM-induced Knowledge distillation framework (KnowSAM) for semi-supervised medical image segmentation. Firstly, we propose a Multi-view Co-training (MC) strategy that employs two distinct sub-networks to employ a co-teaching paradigm, resulting in more robust outcomes. Secondly, we present a Learnable Prompt Strategy (LPS) to dynamically produce dense prompts and integrate an adapter to fine-tune SAM specifically for medical image segmentation tasks. Moreover, we propose SAM-induced Knowledge Distillation (SKD) to transfer useful knowledge from SAM to two sub-networks, enabling them to learn from SAM's predictions and alleviate the effects of incorrect pseudo-labels during training. Notably, the predictions generated by our subnets are used to produce mask prompts for SAM, facilitating effective inter-module information exchange. Extensive experimental results on various medical segmentation tasks demonstrate that our model out-performs the state-of-the-art semi-supervised segmentation approaches. Crucially, our SAM distillation framework can be seamlessly integrated into other semi-supervised segmentation methods to enhance performance. The code will be released upon acceptance of this manuscript at https://***/taozh2017/KnowSAM. © 1982-2012 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Enhancing Deepfake Detection: A Weighted Summation Model of CNN Approach with Local and Global Analysis 2nd

Enhancing Deepfake Detection: A Weighted Summation Model of ...

引用

2nd International Conference on Intelligent Manufacturing and Robotics, ICIMR 2024

作者： Liu, Yuheng Luo, Jiayi Wang, Xiyue Xiao, Hongyan Zhu, Keying Nan, Yanghai Chen, Yi School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou China Co. Ltd. Beijing China School of Intelligent Manufacturing Ecosystem Xi’an Jiaotong-Liverpool University Suzhou China

ISBN: (纸本)9789819639489

In recent years, the advent of sophisticated technologies capable of producing highly realistic images has blurred the line between authenticity and forgery, posing significant societal challenges. The human eye often struggles to discern genuine images from their artificially generated counterparts, necessitating reliable methods to ascertain image authenticity. Scholars have strived to enhance detection precision by refining Convolutional Neural Network (CNN) models. Despite numerous CNN-based models achieving detection accuracies of up to 90%, they are not without their flaws, including limited generalisability, suboptimal parameter solidification, and inadequate scenario-specific refinement, alongside insufficient preprocessing of image data. To address these shortcomings, this research proposes a novel CNN model as a foundation. The image dataset is initially segmented into four equal parts, upon which five independent CNN models are trained, both on these subdivisions and the image in its entirety. The incorporation of a Bayesian model facilitates automatic parameter tuning, followed by a weighted summation of the activation function values derived from the quintet of models, each optimised through training. The aggregated outcome is then evaluated against pre-established criteria to ascertain the veracity of images under various conditions. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

MixDiT: Accelerating Image Diffusion Transformer Inference With Mixed-Precision MX Quantization

IEEE Computer Architecture Letters

引用

IEEE Computer Architecture Letters 2025年

作者： Kim, Daeun Hwang, Jinwoo Oh, Changhun Park, Jongse Korea Advanced Institute of Science and Technology School of Computing Graduate School of AI Semiconductor Daejeon34141 Korea Republic of

Diffusion Transformer (DiT) has driven significant progress in image generation tasks. However, DiT inferencing is notoriously compute-intensive and incurs long latency even on datacenter-scale GPUs, primarily due to its iterative nature and heavy reliance on GEMM operations inherent to its encoder-based structure. To address the challenge, prior work has explored quantization, but achieving low-precision quantization for DiT inferencing with both high accuracy and substantial speedup remains an open problem. To this end, this paper proposes MixDiT, an algorithm-hardware co-designed acceleration solution that exploits mixed Microscaling (MX) formats to quantize DiT activation values. MixDiTquantizes the DiT activation tensors by selectively applying higher precision to magnitude-based outliers, which produce mixed-precision GEMM operations. To achieve tangible speedup from the mixed-precision arithmetic, we design a MixDiTaccelerator that enables precision-flexible multiplications and efficient MX precision conversions. Our experimental results show that MixDiTdelivers a speedup of 2.10-5.32× over RTX 3090, with no loss in FID. © 2025 IEEE.

关键词： Signal encoding

来源：评论

学校读者我要写书评

暂无评论

STMAE: Spatial Temporal Masked Auto-Encoder for Traffic Forecasting 27th

STMAE: Spatial Temporal Masked Auto-Encoder for Traffic For...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Wu, Xing Cai, Chengyou Wang, Xiaoxiao Wang, Jianjia Yao, Junfeng Qian, Quan Song, Jun School of Computer Engineering and Science Shanghai University Shanghai China Shanghai Institute for Advanced Communication and Data Science Shanghai University Shanghai China Shanghai Shentong Metro Co. Ltd. Shanghai China School of AI and Advanced Computing XJTLU Entrepreneur College Xi’an Jiaotong-Liverpool University Suzhou China CSSC Seago System Technology Co. Ltd. Shanghai China Department of Geography Hong Kong Baptist University Hong Kong

ISBN: (纸本)9783031781681

The advancement of intelligent transportation systems underscores the importance of data-driven approaches in traffic forecasting, which plays a crucial role in tasks such as traffic signal control and route guidance, among others. However, the inherent uncertainty stemming from regional traffic dynamics, coupled with intricate spatio-temporal correlations, poses formidable challenges to accurate traffic prediction. Moreover, the complexities inherent in sequence forecasting across varying scales further exacerbate the accuracy dilemma. Recognizing the need for integrating information across spatial and temporal dimensions to enhance forecasting precision, a novel solution termed Spatial Temporal Masked Autoencoder (STMAE) is introduced. The STMAE framework addresses these challenges through a two-stage learning process. In the pre-training phase, an autoencoder architecture is employed to extract spatio-temporal features from the data. In the fine-tuning phase, the pre-trained encoder of the STMAE model undergoes further refinement to specifically target traffic forecasting tasks. Extensive evaluations validate the effectiveness of the proposed STMAE model. Notably, STMAE demonstrates competitive performance, achieving 3.32 Vehs MAE for long-term (60 min) traffic forecasting while operating within a reduced computational budget. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：