Electric vehicle (EV) is a fast-growing technology that could help reduce greenhouse gas emissions in the energy and transportation sectors. The characteristics of EVs allow for rapid smart city expansion and serve as...
详细信息
Recent advancements in computing speed and capacity of Artificial Intelligence (AI) algorithms have reached a saturation level in performance due to the continuous application of Moore's law which resulted in the ...
详细信息
The widely used ReLU is favored for its hardware efficiency, as the implementation at inference is a one bit sign case, yet suffers from issues such as the "dying ReLU" problem, where during training, neuron...
详细信息
Graph-based methods, pivotal for label inference over interconnected objects in many real-world applications, often encounter generalization challenges, if the graph used for model training differs significantly from ...
详细信息
Graph-based methods, pivotal for label inference over interconnected objects in many real-world applications, often encounter generalization challenges, if the graph used for model training differs significantly from the graph used for testing. This work delves into Graph Domain Adaptation (GDA) to address the unique complexities of distribution shifts over graph data, where interconnected data points experience shifts in features, labels, and in particular, connecting patterns. We propose a novel, theoretically principled method, Pairwise Alignment (Pair-Align) to counter graph structure shift by mitigating conditional structure shift (CSS) and label shift (LS). Pair-Align uses edge weights to recalibrate the influence among neighboring nodes to handle CSS and adjusts the classification loss with label weights to handle LS. Our method demonstrates superior performance in real-world applications, including node classification with region shift in social networks, and the pileup mitigation task in particle colliding experiments. For the first application, we also curate the largest dataset by far for GDA studies. Our method shows strong performance in synthetic and other existing benchmark datasets. Copyright 2024 by the author(s)
This research presents AMRes (Adaptive Windowed Convolutional Neural Network and Multiple Residual Network), a novel method that demonstrates remarkable resilience against overfitting, even in scenarios with limited t...
详细信息
High-performance, high conversion ratio power electronics are necessary to enable robotics in future space exploration. This work utilizes the flying capacitor multilevel topology to build a lightweight, efficient con...
详细信息
Drug-drug interaction(DDI)prediction is a crucial issue in molecular *** methods of observing drug-drug interactions through medical experiments require significant resources and *** authors present a Medical Knowledg...
详细信息
Drug-drug interaction(DDI)prediction is a crucial issue in molecular *** methods of observing drug-drug interactions through medical experiments require significant resources and *** authors present a Medical Knowledge Graph Question Answering(MedKGQA)model,dubbed MedKGQA,that predicts DDI by employing machine reading comprehension(MRC)from closed-domain literature and constructing a knowledge graph of“drug-protein”triplets from open-domain *** model vectorises the drug-protein target attributes in the graph using entity embeddings and establishes directed connections between drug and protein entities based on the metabolic interaction pathways of protein targets in the human *** aligns multiple external knowledge and applies it to learn the graph neural *** bells and whistles,the proposed model achieved a 4.5%improvement in terms of DDI prediction accuracy compared to previous state-of-the-art models on the QAngaroo MedHop *** results demonstrate the efficiency and effectiveness of the model and verify the feasibility of integrating external knowledge in MRC tasks.
It's been 10 years since the IEEE launched the Rebooting Computing Initiative (RCI) with the intention to re-examine all levels of how we compute. Back then, the term 'post-Moore' was just coming into vogu...
详细信息
Frequency information (e.g., Discrete Wavelet Transform and Fast Fourier Transform) has been widely applied to solve the issue of Low-Light Image Enhancement (LLIE). However, existing frequency-based models primarily ...
详细信息
This paper introduces a 5G multi-frequency antenna design method based on multi-objective sequential domain patching. By etching helical metamaterials on radiation patches and loading asymmetric electric-inductive-cap...
详细信息
暂无评论