UID:
almafu_9961806445202883
Umfang:
1 online resource (528 pages)
Ausgabe:
1st ed. 2025.
ISBN:
9783031781834
,
303178183X
Serie:
Lecture Notes in Computer Science, 15307
Inhalt:
The multi-volume set of LNCS books with volume numbers 15301-15333 constitutes the refereed proceedings of the 27th International Conference on Pattern Recognition, ICPR 2024, held in Kolkata, India, during December 1–5, 2024. The 963 papers presented in these proceedings were carefully reviewed and selected from a total of 2106 submissions. They deal with topics such as Pattern Recognition; Artificial Intelligence; Machine Learning; Computer Vision; Robot Vision; Machine Vision; Image Processing; Speech Processing; Signal Processing; Video Processing; Biometrics; Human-Computer Interaction (HCI); Document Analysis; Document Recognition; Biomedical Imaging; Bioinformatics.
Anmerkung:
Intro -- President's Address -- Preface -- Organization -- Contents - Part VII -- Graph Matching Networks Meet Optimum-Path Forest: How to Prune Ensembles Efficiently -- 1 Introduction -- 2 Related Works -- 3 Theoretical Background -- 3.1 Optimum-Path Forest -- 3.2 Graph Matching Network -- 4 Proposed Approach -- 5 Methodology -- 5.1 Dataset -- 5.2 Experimental Setup -- 6 Results and Discussions -- 7 Conclusions -- References -- Understanding the Influence of Extremely High-Degree Nodes on Graph Anomaly Detection -- 1 Introduction -- 2 Related Works -- 3 Data Collection and Properties -- 4 Exploring the Influence of Extremely High-Degree Node -- 4.1 Definition of Extremely High-Degree Node -- 4.2 Experimental Settings -- 4.3 Influence of SN on GNN-Based and Non-GNN-Based Models -- 4.4 Impact of SN on Unsupervised and Supervised GADs -- 4.5 Computational Cost -- 4.6 Over-Smoothing -- 5 Method and Experiments -- 5.1 SNGNN -- 5.2 Experiments -- 5.3 Ablation Study -- 6 Conclusion and Limitation -- References -- Spatio-Temporal Heterogeneous Graph Neural Network With Multi-view Learning For Traffic Prediction -- 1 Introduction -- 2 Related Work -- 2.1 Traffic Forecasting -- 2.2 Graph Neural Networks -- 2.3 Dynamic Graph Learning -- 3 Problem Definition -- 4 Methodology -- 4.1 Heterogeneous Temporal Convolution Module -- 4.2 Dynamic Graph Learning Module -- 4.3 Dynamic Graph Convolution Module -- 4.4 Multi-view Fusion Module -- 4.5 Output Module -- 5 Experimental Studies -- 5.1 Datasets and Evaluation Metrics -- 5.2 Experimental Settings -- 5.3 Baselines -- 5.4 Comparison Results -- 5.5 Ablation Study -- 5.6 Effects of Multi-view and Dynamic Graph Learning -- 6 Conclusion -- References -- BotSCL: Heterophily-Aware Social Bot Detection with Supervised Contrastive Learning -- 1 Introduction -- 2 Related Work -- 2.1 Graph-Based Social Bot Detection.
,
2.2 GNNs for Graphs with Heterophily -- 2.3 Contrastive Learning -- 3 Methodology -- 3.1 Graph Augmentation -- 3.2 Aggregation Strategy -- 3.3 Supervised Contrastive Optimization -- 4 Experiments -- 4.1 Experiment Setup -- 4.2 Heterophily Evidence and Influence -- 4.3 Performance Comparison -- 4.4 Ablation Study -- 4.5 Sensitive Analysis -- 4.6 Visualization -- 5 Conclusion -- References -- SimDrop: Towards Deep Graph Convolutional Networks -- 1 Introducation -- 2 Related Work -- 3 Preliminaries -- 4 Method -- 5 Experiment -- 5.1 Experiment Setting -- 5.2 Experimental Results and Analysis -- 6 Conclusion -- References -- A Quantum-inspired Approach to Estimate Optimum-Path Forest Prototypes based on the Traveling Salesman Problem -- 1 Introduction -- 2 Theoretical Background -- 2.1 Optimum-Path Forest -- 2.2 Quantum Machine Learning -- 3 Quantum-inspired Prototype Computation -- 4 Methodology -- 4.1 Datasets -- 4.2 Experimental Setup -- 5 Experiments and Results -- 5.1 Convergence Analysis -- 5.2 Discussion -- 6 Conclusions -- 6.1 Challenges -- 6.2 Future Works -- References -- Face to Cartoon Incremental Super-Resolution Using Knowledge Distillation -- 1 Introduction -- 2 Related Work -- 2.1 Incremental Learning and Knowledge Distillation -- 2.2 Face Super-Resolution -- 3 Proposed Methodology -- 3.1 Problem Description -- 3.2 Knowledge Distillation -- 3.3 Edge Block -- 3.4 Generator Architecture -- 3.5 Discriminator Architecture -- 3.6 Objective Function -- 4 Experimental Settings -- 5 Experimental Results and Discussion -- 5.1 Quantitative Results -- 5.2 Qualitative Results -- 5.3 Ablation Study on Loss Hyperparameters -- 5.4 Cross-Dataset Analysis -- 5.5 Comparsion with Joint Training Approach -- 5.6 Performance on Extended Network -- 6 Conclusion -- References -- Copula Entropy Based Causal Network Discovery from Non-stationary Time Series.
,
1 Introduction -- 2 Related Work -- 3 Preliminary Knowledge -- 3.1 Mutual Information and Transfer Entropy -- 3.2 Conditional Transfer Entropy Estimation -- 3.3 Definitions and Assumptions in Non-stationary Time Series Causal Discovery -- 4 Algorithm Introduction -- 4.1 Causal Structure Learning of the Non-stationary Time Series -- 4.2 Estimation of the Conditional Transfer Entropy -- 4.3 Time Complexity Analysis -- 5 Experimental Results and Analysis -- 5.1 Experiment 1 -- 5.2 Experiment 2 -- 5.3 Experiment 3 -- 5.4 Experimental Analysis -- 6 Real Data Experiment -- 7 Conclusion -- References -- DSparsE: Dynamic Sparse Embedding for Knowledge Graph Completion -- 1 Introduction -- 2 Background and Related Works -- 3 DSparsE for Link Prediction -- 3.1 Dynamic Layer -- 3.2 Relation-Aware Layer -- 3.3 Projection Layer -- 3.4 Residual Layer -- 3.5 Sparse Structure of MLP -- 4 Experiments and Analysis -- 4.1 Datasets and Evaluation Settings -- 4.2 Prediction Performance -- 4.3 Ablation Studies and Further Experiments -- 5 Conclusion -- References -- Interpreting Convolutional Neural Network Decision via Pixel-Wise Interaction Hierarchy Graph -- 1 Introduction -- 2 Related Work -- 3 Method -- 3.1 Interaction Value -- 3.2 Interaction Hierarchy Graph -- 3.3 Filter Contribution -- 4 Experiments -- 4.1 Experimental Settings -- 4.2 Visual Knowledge of Filters -- 4.3 Quantitative Evaluation -- 4.4 Faithfulness Evaluation -- 5 Conclusion -- References -- Denoising Optimization-Based Counterfactual Explanations for Time Series Classification -- 1 Introduction -- 2 Preliminaries -- 2.1 Problem Formalism -- 2.2 Wachter's Method -- 2.3 Contrastive Explanation Method -- 2.4 Counterfactual Explanations Guided by Prototypes -- 2.5 TimeX -- 2.6 Discrete Fourier Transform -- 3 Proposed Approach -- 3.1 Motivation -- 3.2 Low-Pass Filtering -- 3.3 Frequency Clipping.
,
3.4 Time Complexity -- 4 Experimental Setup -- 4.1 Black-Box Classification Models -- 4.2 Datasets -- 4.3 Implementation Details -- 5 Experimental Results -- 5.1 Plausibility -- 5.2 Proximity -- 5.3 Visual Plausibility -- 5.4 PCA -- 5.5 Sensitivity Analysis -- 6 Conclusion -- References -- Improving Adaptive Runoff Forecasts in Data-Scarce Watersheds Through Personalized Federated Learning -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 Problem Formulation -- 3.2 FedHydroDSW: Federated Runoff Forecast for Data-Scarce Watershed -- 4 Empirical Analysis -- 4.1 Dataset and Preprocessing -- 4.2 Design of the Experimental Study -- 4.3 Experimental Result and Discussion -- 4.4 Ablation Study -- 5 Discussion -- 6 Conclusion -- References -- Stagger-Cache MITM: A Privacy-Preserving Hierarchical Model Aggregation Framework -- 1 Introduction -- 2 Related Work -- 3 Contributions of This Paper -- 4 Problem Setup -- 4.1 Mathematical Notation -- 4.2 Performance Criteria -- 5 Multi-tier Inference Frameworks -- 5.1 Framework A: Bottom-Up -- 5.2 Computation Methodology -- 5.3 Framework B: Tier-Caching -- 5.4 Computation Methodology -- 6 Framework of Meet-In-The-Middle (MITM) Staggered Caching -- 6.1 Computation Methodology -- 6.2 Benefits and Drawbacks -- 7 Experimental Results -- 7.1 Use Case: Crop Disease Forecasting -- 7.2 Results: Inference Latency -- 7.3 Results: Accuracy -- 7.4 Results: Memory Used -- 7.5 Results: Data Transmitted over Network -- 7.6 Experiments Discussion -- 8 Conclusion -- References -- ViT2 - Pre-training Vision Transformers for Visual Times Series Forecasting -- 1 Introduction -- 2 Related Work -- 3 Background -- 3.1 Gramian Angular Fields -- 3.2 Vision Transformer -- 4 Methodology -- 4.1 Module 1 - Data Preprocessing -- 4.2 Module 2 - Visualizing Time Series -- 4.3 Module 3 - Probabilistic Forecasting ViT.
,
4.4 Module 4 - Transfer Learning & -- Fine-Tuning -- 5 Experiments and Discussion -- 5.1 Datasets -- 5.2 Experimental Setup -- 5.3 Hyper-parameter Tuning -- 5.4 Results -- 6 Conclusion & -- Future Work -- References -- waLLMartCache: A Distributed, Multi-tenant and Enhanced Semantic Caching System for LLMs -- 1 Introduction -- 2 Related Work -- 3 GPTCache in a Nutshell -- 3.1 Adapter -- 3.2 Pre-processor -- 3.3 Embedding Generator -- 3.4 Cache Manager -- 3.5 Similarity Evaluator -- 3.6 Post-processor -- 4 waLLMartCache: An Enhanced Cache for LLMs -- 4.1 Incorporating Redis as a Database -- 4.2 Designing a Distributed Cache for LLMs -- 4.3 Integrating Multi-tenancy -- 4.4 Improved Decision Engine for Caching -- 4.5 Pre-loading (non-volatile) FAQs into the Cache -- 4.6 Ablation Study -- 5 Conclusion -- References -- ReeSPOT: Reeb Graph Models Semantic Patterns of Normalcy in Human Trajectories -- 1 Introduction -- 2 Methodology -- 2.1 Previous work on Reeb graphs -- 2.2 Reeb graph models agent pattern of normalcy -- 2.3 Construction of Reeb graphs and analysis of time complexity -- 3 Experimentation/Case Study -- 3.1 Data generation -- 3.2 Definition of anomalous behavior -- 3.3 Reeb Graph Generation -- 3.4 Analysis and interpretation of scenarios using Reeb graphs -- 3.5 Reeb graph iteratively detects anomalous behavior of an agent -- 3.6 Quantifying the distance between Reeb graphs -- 3.7 Scalability with Reeb Graphs -- 4 Discussion and Future Work -- References -- Label Disambiguation-Based Feature Selection for Partial Multi-label Learning -- 1 Introduction -- 2 Related Work -- 3 Proposed Approach -- 3.1 Label Disambiguation Using Granular Ball -- 3.2 Feature Selection Using Labeling Confidence -- 4 Experiments -- 4.1 Datasets -- 4.2 Baselines -- 4.3 Experimental Results -- 4.4 Ablation Study -- 5 Conclusion -- References.
,
Neural Encoding of Odors: Translating Odors into Unique Digital Representation with EEG Signals.
Weitere Ausg.:
ISBN 9783031781827
Weitere Ausg.:
ISBN 3031781821
Sprache:
Englisch
DOI:
10.1007/978-3-031-78183-4
Bookmarklink
