다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

Deep learning has revolutionized many machine learning tasks in recent years, ranging from image classification and video processing to speech recognition and natural language understanding. The data in these tasks are typically represented in the Euclidean space. However, there is an increasing number of applications, where data are generated from non-Euclidean domains and are represented as graphs with complex relationships and interdependency between objects. The complexity of graph data has imposed significant challenges on the existing machine learning algorithms. Recently, many studies on extending deep learning approaches for graph data have emerged. In this article, we provide a comprehensive overview of graph neural networks (GNNs) in data mining and machine learning fields. We propose a new taxonomy to divide the state-of-the-art GNNs into four categories, namely, recurrent GNNs, convolutional GNNs, graph autoencoders, and spatial–temporal GNNs. We further discuss the applications of GNNs across various domains and summarize the open-source codes, benchmark data sets, and model evaluation of GNNs. Finally, we propose potential research directions in this rapidly growing field.

/pdf/a-comprehensive-survey-on-graph-neural-networks-4lw9i4v183.pdf

A Comprehensive Survey on Graph Neural Networks

Lots of learning tasks require dealing with graph data which contains rich relation information among elements. Modeling physics systems, learning molecular fingerprints, predicting protein interface, and classifying diseases demand a model to learn from graph inputs. In other domains such as learning from non-structural data like texts and images, reasoning on extracted structures (like the dependency trees of sentences and the scene graphs of images) is an important research topic which also needs graph reasoning models. Graph neural networks (GNNs) are neural models that capture the dependence of graphs via message passing between the nodes of graphs. In recent years, variants of GNNs such as graph convolutional network (GCN), graph attention network (GAT), graph recurrent network (GRN) have demonstrated ground-breaking performances on many deep learning tasks. In this survey, we propose a general design pipeline for GNN models and discuss the variants of each component, systematically categorize the applications, and propose four open problems for future research.

Graph Neural Networks: A Review of Methods and Applications

Learning vector representations (aka. embeddings) of users and items lies at the core of modern recommender systems. Ranging from early matrix factorization to recently emerged deep learning based methods, existing efforts typically obtain a user's (or an item's) embedding by mapping from pre-existing features that describe the user (or the item), such as ID and attributes. We argue that an inherent drawback of such methods is that, the collaborative signal, which is latent in user-item interactions, is not encoded in the embedding process. As such, the resultant embeddings may not be sufficient to capture the collaborative filtering effect. In this work, we propose to integrate the user-item interactions - more specifically the bipartite graph structure - into the embedding process. We develop a new recommendation framework Neural Graph Collaborative Filtering (NGCF), which exploits the user-item graph structure by propagating embeddings on it. This leads to the expressive modeling of high-order connectivity in user-item graph, effectively injecting the collaborative signal into the embedding process in an explicit manner. We conduct extensive experiments on three public benchmarks, demonstrating significant improvements over several state-of-the-art models like HOP-Rec [39] and Collaborative Memory Network [5]. Further analysis verifies the importance of embedding propagation for learning better user and item representations, justifying the rationality and effectiveness of NGCF. Codes are available at https://github.com/xiangwang1223/neural_graph_collaborative_filtering.

Neural Graph Collaborative Filtering

Recent advancements in deep neural networks for graph-structured data have led to state-of-the-art performance on recommender system benchmarks. However, making these methods practical and scalable to web-scale recommendation tasks with billions of items and hundreds of millions of users remains an unsolved challenge. Here we describe a large-scale deep recommendation engine that we developed and deployed at Pinterest. We develop a data-efficient Graph Convolutional Network (GCN) algorithm, which combines efficient random walks and graph convolutions to generate embeddings of nodes (i.e., items) that incorporate both graph structure as well as node feature information. Compared to prior GCN approaches, we develop a novel method based on highly efficient random walks to structure the convolutions and design a novel training strategy that relies on harder-and-harder training examples to improve robustness and convergence of the model. We also develop an efficient MapReduce model inference algorithm to generate embeddings using a trained model. Overall, we can train on and embed graphs that are four orders of magnitude larger than typical GCN implementations. We show how GCN embeddings can be used to make high-quality recommendations in various settings at Pinterest, which has a massive underlying graph with 3 billion nodes representing pins and boards, and 17 billion edges. According to offline metrics, user studies, as well as A/B tests, our approach generates higher-quality recommendations than comparable deep learning based systems. To our knowledge, this is by far the largest application of deep graph embeddings to date and paves the way for a new generation of web-scale recommender systems based on graph convolutional architectures.

Graph Convolutional Neural Networks for Web-Scale Recommender Systems

Recent advancements in deep neural networks for graph-structured data have led to state-of-the-art performance on recommender system benchmarks. However, making these methods practical and scalable to web-scale recommendation tasks with billions of items and hundreds of millions of users remains a challenge. Here we describe a large-scale deep recommendation engine that we developed and deployed at Pinterest. We develop a data-efficient Graph Convolutional Network (GCN) algorithm PinSage, which combines efficient random walks and graph convolutions to generate embeddings of nodes (i.e., items) that incorporate both graph structure as well as node feature information. Compared to prior GCN approaches, we develop a novel method based on highly efficient random walks to structure the convolutions and design a novel training strategy that relies on harder-and-harder training examples to improve robustness and convergence of the model. We also develop an efficient MapReduce model inference algorithm to generate embeddings using a trained model. We deploy PinSage at Pinterest and train it on 7.5 billion examples on a graph with 3 billion nodes representing pins and boards, and 18 billion edges. According to offline metrics, user studies and A/B tests, PinSage generates higher-quality recommendations than comparable deep learning and graph-based alternatives. To our knowledge, this is the largest application of deep graph embeddings to date and paves the way for a new generation of web-scale recommender systems based on graph convolutional architectures.

Graph Convolutional Neural Networks for Web-Scale Recommender Systems.

The heating of a solar cell has significant adverse consequences on both its efficiency and its reliability. Here to fully exploit the cooling potential of solar cells, we experimentally characterized the thermal radiation and solar absorption properties of current silicon solar cells and, on the basis of such experimental characterization, propose a comprehensive photonic approach by simultaneously performing radiative cooling while also selectively utilizing sunlight. In particular, we design a photonic cooler made of one-dimensional photonic films that can strongly radiate heat through its thermal emission while also significantly reflecting the solar spectrum in the sub-band-gap and ultraviolet regimes. We show that applying this photonic cooler to a solar panel can lower the cell temperature by over 5.7 °C. We also show that this photonic cooler can be used in a concentrated photovoltaic system to significantly reduce the solar cell temperature or required cooling power. This photonic cooler can be r...

A Comprehensive Photonic Approach for Solar Cell Cooling

Lithium metal is the ultimate anode choice for high energy density rechargeable lithium batteries. However, it suffers from inferior electrochemical performance and safety issues due to its high reactivity and the growth of lithium dendrites. It has long been desired to develop a materials coating on Li metal, which is pinhole-free, mechanically robust without fracture during Li metal deposition and stripping, and chemically stable against Li metal and liquid electrolytes, all while maintaining adequate ionic conductivity. However, such an ideal material coating has yet to be found. Here we report a novel synthesis method by reacting clean molten lithium foil directly with pure nitrogen gas to generate instantaneously a pinhole-free and ionically conductive α-Li3N film directly bonded onto Li metal foil. The film consists of highly textured large Li3N grains (tens of μm) with (001) crystalline planes parallel to the Li metal surface. The bonding between textured grains is strong, resulting in a mechanical...

Robust Pinhole-free Li3N Solid Electrolyte Grown from Molten Lithium.

With the ability of harvesting the coldness of universe as a thermodynamic resource, radiative cooling technology is important for a broad range of applications such as passive building cooling, refrigeration, and renewable energy harvesting. However, all existing radiative cooling technologies utilize static structures, which lack the ability of self-adaptive tuning based on demand. Here we present the concept of self-adaptive radiative cooling based on phase change materials such as vanadium dioxide. We design a photonic structure that can adaptively turn 'on' and 'off' radiative cooling, depending the ambient temperature, without any extra energy input for switching. Our results here lead to new functionalities of radiative cooling and can potentially be used in a wide range of applications for the thermal managements of buildings, vehicles and textiles.

Kaifeng Chen

Papers

Graph Convolutional Neural Networks for Web-Scale Recommender Systems

Graph Convolutional Neural Networks for Web-Scale Recommender Systems.

A Comprehensive Photonic Approach for Solar Cell Cooling

Robust Pinhole-free Li3N Solid Electrolyte Grown from Molten Lithium.

Self-adaptive radiative cooling based on phase change materials