检索结果-内蒙古大学图书馆

Benchmarking Graph Representations and Graph Neural Networks for Multivariate Time Series Classification

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Yang, Wennuo Wu, Shiling Zhou, Yuzhi Luo, Cheng He, Xilin Xie, Weicheng Shen, Linlin Song, Siyang Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University China Shenzhen Institute of Artificial Intelligence and Robotics for Society China Guangdong Provincial Key Laboratory of Intelligent Information Processing China HBUG Lab University of Exeter United Kingdom

Multivariate Time Series Classification (MTSC) enables the analysis if complex temporal data, and thus serves as a cornerstone in various real-world applications, ranging from healthcare to finance. Since the relationship among variables in MTS usually contain crucial cues, a large number of graph-based MTSC approaches have been proposed, as the graph topology and edges can explicitly represent relationships among variables (channels), where not only various MTS graph representation learning strategies but also different Graph Neural Networks (GNNs) have been explored. Despite such progresses, there is no comprehensive study that fairly benchmarks and investigates the performances of existing widely-used graph representation learning strategies/GNN classifiers in the application of different MTSC tasks. In this paper, we present the first benchmark which systematically investigates the effectiveness of the widely-used three node feature definition strategies, four edge feature learning strategies and five GNN architecture, resulting in 60 different variants for graph-based MTSC. These variants are developed and evaluated with a standardized data pipeline and training/validation/testing strategy on 26 widely-used suspensor MTSC datasets. Our experiments highlight that node features significantly influence MTSC performance, while the visualization of edge features illustrates why adaptive edge learning outperforms other edge feature learning methods. The code of the proposed benchmark is publicly available at https://***/CVI-yangwn/*** Codes 68T10 © 2025, CC BY.

关键词： Graph neural networks

CATransUnetLBP: Accurate Prediction of Protein-Ligand Binding Pockets Using a Hybrid Network

学校读者我要写书评

暂无评论

IEEE Transactions on Computational Biology and Bioinformatic...

IEEE Transactions on Computational Biology and Bioinformatics 2025年第1期22卷 355-367页

作者： Cheng Cai Zhaohong Deng Andong Li Yun Zuo Haoran Chen Zhisheng Wei Lei Wang Xiaoyong Pan Hong-Bin Shen Dong-Jun Yu School of Artificial Intelligence and Computer Science Jiangnan University Wuxi China Shanghai Key Lab of Intell. Info. Processing School of CS Fudan University Shanghai China School of Biotechnology Jiangnan University Wuxi China Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

The development of intelligent methods capable of predicting protein-ligand binding sites has become a popular research field. Recently, deep learning based methods have been proposed as a promising solution for this task. However, some limitations still exist. For example, the network structure is not optimized for predicting protein binding pockets, which limits the model's capabilities. To address the aforementioned challenges, a novel method called CATransUnetLPB is proposed, in which a new network structure named CATransUnet is designed. The proposed CATransUnet combines CNN and Transformer models to accurately segment binding pocket regions from protein 3D structures. It outperforms existing representative methods on three test sets, demonstrating the effectiveness of optimizing the deep network model for detecting protein ligand binding pockets. Furthermore, we conduct thorough analysis on applying data augmentation to protein data structure and confirm that such technique can enhance the model's generalization ability, thereby ensuring good performance on new protein structures. Moreover, experiments show that the predicted binding pockets from our model can complement the results obtained from other methods. This suggests that integrating our method with existing approaches could further improve the prediction of protein-ligand binding pockets.

关键词： Proteins Three-dimensional displays Probes Vectors Image segmentation Deep learning Atoms Surface treatment Prediction algorithms Training

Molecule generation for drug design: A graph learning perspective

学校读者我要写书评

暂无评论

Fundamental Research 2025年

作者： Yang, Nianzu Wu, Huaijin Zeng, Kaipeng Li, Yang Bao, Siyuan Yan, Junchi School of Artificial Intelligence & Department of Computer Science and Engineering & MoE Lab of AI Shanghai Jiao Tong University Shanghai 200240 China Shanghai Pinghe School Shanghai 201208 China

Machine learning, particularly graph learning, is gaining increasing recognition for its transformative impact across various fields. One such promising application is in the realm of molecule design and discovery, notably within the pharmaceutical industry. Our survey offers a comprehensive overview of state-of-the-art methods in molecule design, particularly focusing on de novo drug design, which incorporates (deep) graph learning techniques. We categorize these methods into three distinct groups: i) all-at-once, ii) fragment-based, and iii) node-by-node. Additionally, we introduce some key public datasets and outline the commonly used evaluation metrics for both the generation and optimization of molecules. In the end, we discuss the existing challenges in this field and suggest potential directions for future research. © 2024

关键词： Drug discovery Generative models Graph generation Graph representation learning Machine learning

Generalizable Audio Deepfake Detection via Latent Space Refinement and Augmentation

学校读者我要写书评

暂无评论

Generalizable Audio Deepfake Detection via Latent Space Refi...

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Wen Huang Yanmei Gu Zhiming Wang Huijia Zhu Yanmin Qian AI Institute Department of Computer Science and Engineering Auditory Cognition and Computational Acoustics Lab MoE Key Lab of Artificial Intelligence Shanghai Jiao Tong University Shanghai China SJTU Paris Elite Institute of Technology Ant Group Shanghai China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Advances in speech synthesis technologies, like text-to-speech (TTS) and voice conversion (VC), have made detecting deepfake speech increasingly challenging. Spoofing countermeasures often struggle to generalize effectively, particularly when faced with unseen attacks. To address this, we propose a novel strategy that integrates Latent Space Refinement (LSR) and Latent Space Augmentation (LSA) to improve the generalization of deepfake detection systems. LSR introduces multiple learnable prototypes for the spoof class, refining the latent space to better capture the intricate variations within spoofed data. LSA further diversifies spoofed data representations by applying augmentation techniques directly in the latent space, enabling the model to learn a broader range of spoofing patterns. We evaluated our approach on four representative datasets, i.e. ASVspoof 2019 LA, ASVspoof 2021 LA and DF, and In-The-Wild. The results show that LSR and LSA perform well individually, and their integration achieves competitive results, matching or surpassing current state-of-the-art methods.

关键词： Deepfakes Refining Prototypes Data visualization Signal processing Robustness Data models Acoustics Text to speech

Improving the Sparse Structure Learning of Spiking Neural Networks from the View of Compression Efficiency

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Shen, Jiangrong Xu, Qi Pan, Gang Chen, Badong Faculty of Electronic and Information Engineering Xi’an Jiaotong University China Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University China State Key Lab of Brain-Machine Intelligence Zhejiang University China National Key Lab of Human-Machine Hybrid Augmented Intelligence Xi’an Jiaotong University China School of Computer Science Dalian University of Technology China

The human brain utilizes spikes for information transmission and dynamically reorganizes its network structure to boost energy efficiency and cognitive capabilities throughout its lifespan. Drawing inspiration from this spike-based computation, Spiking Neural Networks (SNNs) have been developed to construct event-driven models that emulate this efficiency. Despite these advances, deep SNNs continue to suffer from over-parameterization during training and inference, a stark contrast to the brain’s ability to self-organize. Furthermore, existing sparse SNNs are challenged by maintaining optimal pruning levels due to a static pruning ratio, resulting in either under or over-pruning. In this paper, we propose a novel two-stage dynamic structure learning approach for deep SNNs, aimed at maintaining effective sparse training from scratch while optimizing compression efficiency. The first stage evaluates the compressibility of existing sparse subnetworks within SNNs using the PQ index, which facilitates an adaptive determination of the rewiring ratio for synaptic connections based on data compression insights. In the second stage, this rewiring ratio critically informs the dynamic synaptic connection rewiring process, including both pruning and regrowth. This approach significantly improves the exploration of sparse structures training in deep SNNs, adapting sparsity dynamically from the point view of compression efficiency. Our experiments demonstrate that this sparse training approach not only aligns with the performance of current deep SNNs models but also significantly improves the efficiency of compressing sparse SNNs. Crucially, it preserves the advantages of initiating training with sparse models and offers a promising solution for implementing Edge AI on neuromorphic hardware. © 2025, CC BY-NC-ND.

关键词：

Utilization of DenseNet201, EfficientNetB3, Resnet50, and VGG19 as Pre-Trained Convolutional Neural Network Models for Brain Tumour Classification

学校读者我要写书评

暂无评论

Utilization of DenseNet201, EfficientNetB3, Resnet50, and VG...

International Conference on artificial intelligence in Information and Communication (ICAIIC)

作者： Kanwarpartap Singh Gill Rupesh Gupta Gunjan Shandilya Srinivas Aluvala Adel Sulaiman Hani Alshahrani Mana Saleh Al Reshan Asadullah Shaikh Chitkara University Institute of Engineering and Technology Chitkara University Punjab India School of Computer Science & Artificial Intelligence SR University Warangal India Department of Computer Science College of Computer Science and Information Systems Najran University Najran Saudi Arabia Emerging Technologies Research Lab (ETRL) College of Computer Science and Information Systems Najran University Najran Saudi Arabia Department of Information Systems College of Computer Science and Information Systems Najran University Najran Saudi Arabia

ISBN: (数字)9798331507022

ISBN: (纸本)9798331507039

A major development in medical imaging diagnosis is the use of pretrained Convolutional Neural Network (CNN) models for brain tumour classification. This study uses pretrained CNN EfficientNetB3, ResNet50, and VGG19, to explore the crucial field of early brain tumour identification. The dataset used in the research consists of 7023 brain MRI images, which are divided into 2000 validation images, 1311 testing images, and 3712 training images. After preparing the data, we want to see which CNN designs work best for sorting images of gliomas, meningiomas, no tumors, and pituitaries into different groups. The study shows how important it is to find brain tumors early and how we can make brain imaging tests more accurate and helpful for diagnosing them. The EfficientNetB3 model is the best and wins with remarkable 99% accuracy. The results show the good and bad points of different models and give important information to help improve skills in diagnosing using neuroimaging.

关键词： Neuroimaging Training Accuracy Magnetic resonance imaging Brain tumors Brain modeling Convolutional neural networks Medical diagnostic imaging Residual neural networks Testing

Structural and Statistical Texture Knowledge Distillation and Learning for Segmentation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Ji, Deyi Zhao, Feng Lu, Hongtao Wu, Feng Ye, Jieping MoE Key Laboratory of Brain-inspired Intelligent Perception and Cognition University of Science and Technology of China China Department of Computer Science and Engineering MOE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University China Alibaba Group China

Low-level texture feature/knowledge is also of vital importance for characterizing the local structural pattern and global statistical properties, such as boundary, smoothness, regularity, and color contrast, which may not be well addressed by high-level deep features. In this paper, we aim to re-emphasize the low-level texture information in deep networks for semantic segmentation and related knowledge distillation tasks. To this end, we take full advantage of both structural and statistical texture knowledge and propose a novel Structural and Statistical Texture Knowledge Distillation (SSTKD) framework for semantic segmentation. Specifically, Contourlet Decomposition Module (CDM) is introduced to decompose the low-level features with iterative Laplacian pyramid and directional filter bank to mine the structural texture knowledge, and Texture Intensity Equalization Module (TIEM) is designed to extract and enhance the statistical texture knowledge with the corresponding Quantization Congruence Loss (QDL). Moreover, we propose the Co-occurrence TIEM (C-TIEM) and generic segmentation frameworks, namely STLNet++ and U-SSNet, to enable existing segmentation networks to harvest the structural and statistical texture information more effectively. Extensive experimental results on three segmentation tasks demonstrate the effectiveness of the proposed methods and their state-of-the-art performance on seven popular benchmark datasets, respectively. Copyright © 2025, The Authors. All rights reserved.

关键词： Semantic Segmentation

Generalizable Audio Deepfake Detection via Latent Space Refinement and Augmentation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Huang, Wen Gu, Yanmei Wang, Zhiming Zhu, Huijia Qian, Yanmin Auditory Cognition and Computational Acoustics Lab MoE Key Lab of Artificial Intelligence AI Institute Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China SJTU Paris Elite lnstitute of Technology China Ant Group Shanghai China

关键词： Speech recognition

Edge-Enhanced Cascaded MRF for SAR Image Segmentation

学校读者我要写书评

暂无评论

IEEE Transactions on Geoscience and Remote Sensing 2025年 63卷

作者： Liu, Mengmeng Shang, Ronghua Liu, Kang Feng, Jie Wang, Chao Xu, Songhua Li, Yangyang Xidian University Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education School of Artificial Intelligence Shaanxi Xi’an710071 China Xidian University School of Computer Science and Technology Shaanxi Xi’an710071 China Research Center for Data Hub and Security Zhejiang Lab Hangzhou311100 China The Second Affiliated Hospital of Xi’an Jiaotong University Department of Health Management Institute of Medical Artificial Intelligence Xi’an710004 China

Markov random fields (MRFs) effectively capture local contextual information by modeling the spatial dependencies between pixels, which helps highlight details and enhances segmentation smoothness. To fully exploit MRF for synthetic aperture radar (SAR) image segmentation, we propose a novel edge-enhanced cascaded MRF (ECMRF) approach. Specifically, we introduce multiple edge-constrained filters to emphasize SAR image boundaries and provide relatively clean features. Building on this, we present a cascaded MRF framework that sequentially integrates region-level and pixel-level segmentations with feature perturbation and fusion to generate the final segmentation output. The framework comprises four key components: 1) a region-level MRF (RMRF), regulated by edge features, to achieve precise region segmentation;2) a pixel-level MRF (PMRF) with selective label smoothing to refine edges and reduce noise clusters;3) equal-channel feature perturbation (EC-FP) to increase feature diversity;and 4) a random probability-based feature fusion (RP-FF) scheme to merge the input features. The experimental results demonstrate that our ECMRF outperforms six state-of-the-art comparable methods, underscoring its competitive performance. © 1980-2012 IEEE.

关键词： Image segmentation