ICDE 2024
389 papers
The 2024 IEEE International Conference on Data Engineering (ICDE) featured 583 papers covering a broad range of topics in data management, privacy, and machine learning integration. A primary research direction involved federated learning, specifically addressing model heterogeneity, privacy leakage, and robustness against adversarial model poisoning. Graph neural networks (GNNs) were also prominent, with studies investigating link prediction, community search in heterogeneous networks, and differential privacy applications.
Additional research themes included advancements in natural language processing and semantic analysis, such as hybrid semantic similarity for textual alignment and scaling NLP with distant supervision. The conference also covered temporal data analysis, including differentiable temporal alignment and segmented regression for identifying inflection points. Industrial and practical applications were represented by work on data lake storage and functional dependency discovery in outsourced databases. Prolific contributors to this edition included researchers such as Xiaofang Zhou, Xuemin Lin, and Yunjun Gao.
Top Research Directions
Research directions most represented by ICDE 2024 papers.
Top Authors
Impactful Papers
Papers from ICDE 2024 that shaped research clusters.
Influential
HYPPO: Using Equivalences to Optimize Pipelines in Exploratory Machine Learning
influential
Incremental Execution for Iterative Mach
HeteFedRec: Federated Recommender Systems with Model Heterogeneity
influential
Federated Learning with Diverse Model Ar
Learning Multi-Pattern Normalities in the Frequency Domain for Efficient Time Series Anomaly Detecti
influential
Hyperparameter Sensitivity in Spectral F
MUST: An Effective and Scalable Framework for Multimodal Search of Target Modality
influential
Incremental Maintenance of Dynamic Graph
Generative and Contrastive Paradigms Are Complementary for Graph Self-Supervised Learning
influential
Over-Smoothing in Graph Neural Networks
Quantum Data Management: From Theory to Opportunities
influential
Mapping Optimization Problems to Quadrat
MUST: An Effective and Scalable Framework for Multimodal Search of Target Modality
influential
Content Based Indexing and Retrieval of
LightTR: A Lightweight Framework for Federated Trajectory Recovery
influential
Robust Federated Learning under Statisti
ADecimo: Model Selection for Time Series Anomaly Detection
influential
Predicting Downstream Performance Withou
VDTuner: Automated Performance Tuning for Vector Data Management Systems
influential
Informed Initialization for Bayesian Opt