Export Citations
Save this search
Please login to be able to save your searches and receive alerts for new content matching your search criteria.
- ArticleSeptember 2024
An Accuracy-Shaping Mechanism for Competitive Distributed Learning
Artificial Neural Networks and Machine Learning – ICANN 2024Pages 143–158https://doi.org/10.1007/978-3-031-72347-6_10AbstractIn competitive distributed learning, organizations face the challenge of collaboratively training machine learning models without sharing sensitive raw data, while competing for the same customer base using model-based services. Federated learning ...
- ArticleSeptember 2024
Bridging Dimensions: Confident Reachability for High-Dimensional Controllers
AbstractAutonomous systems are increasingly implemented using end-to-end learning-based controllers. Such controllers make decisions that are executed on the real system, with images as one of the primary sensing modalities. Deep neural networks form a ...
- ArticleSeptember 2024
Reinventing Node-centric Traffic Forecasting for Improved Accuracy and Efficiency
Machine Learning and Knowledge Discovery in Databases. Research TrackPages 21–38https://doi.org/10.1007/978-3-031-70352-2_2AbstractTraffic forecasting is a crucial application in smart city efforts. After revisiting the existing literature on deep learning-based traffic forecasting methods, we identify two primary research approaches: node-centric and graph-centric. Node-...
- research-articleAugust 2024
HiGPT: Heterogeneous Graph Language Model
KDD '24: Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data MiningPages 2842–2853https://doi.org/10.1145/3637528.3671987Heterogeneous graph learning aims to capture complex relationships and diverse relational semantics among entities in a heterogeneous graph to obtain meaningful representations for nodes and edges. Recent advancements in heterogeneous graph neural ...
- research-articleAugust 2024
MSPipe: Efficient Temporal GNN Training via Staleness-Aware Pipeline
KDD '24: Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data MiningPages 2651–2662https://doi.org/10.1145/3637528.3671844Memory-based Temporal Graph Neural Networks (MTGNNs) are a class of temporal graph neural networks that utilize a node memory module to capture and retain long-term temporal dependencies, leading to superior performance compared to memory-less ...
-
- research-articleAugust 2024
UrbanGPT: Spatio-Temporal Large Language Models
KDD '24: Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data MiningPages 5351–5362https://doi.org/10.1145/3637528.3671578Spatio-temporal prediction aims to forecast and gain insights into the ever-changing dynamics of urban environments across both time and space. Its purpose is to anticipate future patterns, trends, and events in diverse facets of urban life, including ...
- tutorialAugust 2024
A Survey of Large Language Models for Graphs
KDD '24: Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data MiningPages 6616–6626https://doi.org/10.1145/3637528.3671460Graphs are an essential data structure utilized to represent relationships in real-world scenarios. Prior research has established that Graph Neural Networks (GNNs) deliver impressive outcomes in graph-centric tasks, such as link prediction and node ...
- research-articleJuly 2024
A wideband inductorless LNA utilizing common-source noise-canceling and cascode configuration
AbstractThe common-source noise-canceling (CSNC) and the gyrator are known for alleviating the trade-off between NF and S11 in wideband LNAs. Moreover, the departure of the input-matching stage and signal-amplifying stage in CSNC facilitates the design ...
- research-articleJuly 2024
SaliencyCut: Augmenting plausible anomalies for anomaly detection
AbstractAnomaly detection under the open-set scenario is a challenging task that requires learning discriminative features to detect anomalies that were even unseen during training. As a cheap yet effective approach, data augmentation has been widely ...
Graphical abstractDisplay Omitted
Highlights- We propose a saliency-guided data augmentation for creating pseudo anomalies, namely SaliencyCut.
- We deploy a two-head learning mechanism based on the theoretical lower bound.
- We further design a novel patch-wise residual module to ...
- research-articleJuly 2024
Recovery of bandlimited graph signals based on the reproducing kernel Hilbert space
AbstractSignal recovery on graphs is attracting more and more attentions. Based on the smoothness assumption, the signal recovery problem can be formulated as an unconstrained optimization model. Although the model can be solved analytically, the ...
- research-articleJuly 2024
GraphGPT: Graph Instruction Tuning for Large Language Models
SIGIR '24: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information RetrievalPages 491–500https://doi.org/10.1145/3626772.3657775Graph Neural Networks (GNNs) have evolved to understand graph structures through recursive exchanges and aggregations among nodes. To enhance robustness, self-supervised learning (SSL) has become a vital tool for data augmentation. Traditional methods ...
- research-articleJuly 2024
SelfGNN: Self-Supervised Graph Neural Networks for Sequential Recommendation
SIGIR '24: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information RetrievalPages 1609–1618https://doi.org/10.1145/3626772.3657716Sequential recommendation effectively addresses information overload by modeling users' temporal and sequential interaction patterns. To overcome the limitations of supervision signals, recent approaches have adopted self-supervised learning techniques ...
- research-articleJuly 2024
A wideband inductorless LNA exploiting three-stage feedback and thermal noise canceling
AbstractIn inductorless wideband low-noise amplifier (LNA) design, input-matching is traded off with noise figure (NF). The common-source noise-canceling (CSNC) structure and the gyrator alleviate this trade-off, but the corresponding study is ...
- research-articleJuly 2024
KaryoXpert: An accurate chromosome segmentation and classification framework for karyotyping analysis without training with manually labeled metaphase-image mask annotations
- Siyuan Chen,
- Kaichuang Zhang,
- Jingdong Hu,
- Na Li,
- Ao Xu,
- Haoyang Li,
- Juexiao Zhou,
- Chao Huang,
- Yongguo Yu,
- Xin Gao
Computers in Biology and Medicine (CBIM), Volume 177, Issue Chttps://doi.org/10.1016/j.compbiomed.2024.108601AbstractAutomated karyotyping is of great importance for cytogenetic research, as it speeds up the process for cytogeneticists through incorporating AI-driven automated segmentation and classification techniques. Existing frameworks confront two primary ...
Highlights- KaryoXpert can be trained without manually annotated ground truth instance masks.
- It merges advantages from both morphology algorithms and deep-learning models.
- It performs high accuracy despite domain shifts, batch effects, and ...
- research-articleJuly 2024
Uncertainty-aware prototypical learning for anomaly detection in medical images
AbstractAnomalous object detection (AOD) in medical images aims to recognize the anomalous lesions, and is crucial for early clinical diagnosis of various cancers. However, it is a difficult task because of two reasons: (1) the diversity of the anomalous ...
- tutorialMay 2024
Large Language Models for Graphs: Progresses and Directions
WWW '24: Companion Proceedings of the ACM Web Conference 2024Pages 1284–1287https://doi.org/10.1145/3589335.3641251Graph neural networks (GNNs) have emerged as fundamental methods for handling structured graph data in various domains, including citation networks, molecule prediction, and recommender systems. They enable the learning of informative node or graph ...
- research-articleMay 2024
GraphPro: Graph Pre-training and Prompt Learning for Recommendation
WWW '24: Proceedings of the ACM Web Conference 2024Pages 3690–3699https://doi.org/10.1145/3589334.3645546GNN-based recommendation systems have been successful in capturing complex user-item interactions using multi-hop message passing. However, these methods often struggle to handle the dynamic nature of user-item interactions, making it challenging to ...
- research-articleMay 2024
Representation Learning with Large Language Models for Recommendation
WWW '24: Proceedings of the ACM Web Conference 2024Pages 3464–3475https://doi.org/10.1145/3589334.3645458Recommender systems have seen significant advancements with the influence of deep learning and graph neural networks, particularly in capturing complex user-item relationships. However, these graph-based recommenders heavily depend on ID-based data, ...
- research-articleMay 2024
PromptMM: Multi-Modal Knowledge Distillation for Recommendation with Prompt-Tuning
WWW '24: Proceedings of the ACM Web Conference 2024Pages 3217–3228https://doi.org/10.1145/3589334.3645359Multimedia online platforms (e.g., Amazon, TikTok) have greatly benefited from the incorporation of multimedia (e.g., visual, textual, and acoustic) content into their personal recommender systems. These modalities provide intuitive semantics that ...
- research-articleJuly 2024
Video surveillance-based multi-task learning with swin transformer for earthwork activity classification
Engineering Applications of Artificial Intelligence (EAAI), Volume 131, Issue Chttps://doi.org/10.1016/j.engappai.2023.107814AbstractBulldozers, pivotal in earthworks, traditionally undergo supervision through labor-intensive and potentially unreliable manual methods. This research proposes a vision-based method for automating the monitoring of bulldozer operations. First, ...
Highlights- An automatic method for earthwork activity classification is proposed.
- Shoveling action, shoveling status and soil classification are detected simultaneously.
- A video multi-task model based on Video Swin Transformer is developed.