Due to the complexity of cancer pathogenesis at different omics levels, it is necessary to find a comprehensive method to accurately distinguish and find cancer subtypes for cancer treatment. In this paper, we propose...
详细信息
Due to the complexity of cancer pathogenesis at different omics levels, it is necessary to find a comprehensive method to accurately distinguish and find cancer subtypes for cancer treatment. In this paper, we proposed a new cancer multi-omics subtype identification method, which is based on variational autoencoder measured by Wasserstein distance and graph autoencoder (WVGMO). This method depends on two foremost models. The first model is a variational autoencoder measured by Wasserstein distance (WVAE), which is used to extract potential spatial information of each omic data type. The second model is the graph autoencoder (GAE) with the second-order proximity. It has the capability to retain the topological structure information and feature information of the multi-omics data. And then, the identification of cancer subtypes via k-means clustering. Extensive experiments were conducted on seven different cancers based on four omics data from TCGA. The results show that WVGMO provides equivalent or even better results than the most of advanced synthesis methods.
It is critical to correctly assemble high-dimensional single-cell RNA sequencing (scRNA-seq) datasets and downscale them for downstream analysis. However, given the complex relationships between cells, it remains a ch...
详细信息
It is critical to correctly assemble high-dimensional single-cell RNA sequencing (scRNA-seq) datasets and downscale them for downstream analysis. However, given the complex relationships between cells, it remains a challenge to simultaneously eliminate batch effects between datasets and maintain the topology between cells within each dataset. Here, we propose scGAMNN, a deep learning model based on graph autoencoder, to simultaneously achieve batch correction and topology-preserving dimensionality reduction. The low-dimensional integrated data obtained by scGAMNN can be used for visualization, clustering and trajectory *** comparing it with the other five methods, multiple tasks show that scGAMNN consistently has comparable data integration performance in clustering and trajectory conservation.
Many complex systems in the real world can be characterized as attributed networks. To mine the poten-tial information in these networks, deep embedded clustering, which obtains node representations and clusters simul...
详细信息
Many complex systems in the real world can be characterized as attributed networks. To mine the poten-tial information in these networks, deep embedded clustering, which obtains node representations and clusters simultaneously, has been given much attention in recent years. Under the assumption of consis-tency for data in different views, the cluster structure of network topology and that of node attributes should be consistent for an attributed network. However, many existing methods ignore this property, even though they separately encode node representations from network topology and node attributes and cluster nodes on representation vectors learned from one of the views. Therefore, in this study, we propose an end-to-end deep embedded clustering model for attributed networks. It utilizes graph autoen-coder and node attribute autoencoder to learn node representations and cluster assignments. In addition, a distribution consistency constraint is introduced to maintain the latent consistency of cluster distri-butions in two views. Extensive experiments on several datasets demonstrate that the proposed model achieves significantly better or competitive performance compared with the state-of-the-art methods. The source code can be found at https://***/Zhengymm/DCP.(c) 2023 Elsevier Ltd. All rights reserved.
This project showcases a use case away from most other research in the field of generative AI in architecture. We present a workflow to generate new, three-dimensional spatial configurations of buildings by sampling t...
详细信息
ISBN:
(纸本)9789887891819
This project showcases a use case away from most other research in the field of generative AI in architecture. We present a workflow to generate new, three-dimensional spatial configurations of buildings by sampling the latent space of a graph autoencoder. graph representations of three-dimensional buildings can store more data and hence reduce the loss of information from building to machine learning model compared to image- and voxel-based representations. graphs do not only represent information about elements (nodes/pixels/etc.) but also the relationships between elements (edges). This is specifically helpful in architecture where we define space as an assemblage of physical elements which are all somehow connected (i.e., wall touches floor). Our method generates valuable, logical and original geometries that represent the architectural style chosen in the training data. These geometries are highly different from any image-based generation process and justify the importance of graph-based 3D geometry generation of architecture via machine learning. The method also introduces a novel conversion process from architecture to graph, an adapted decoder architecture, and a physical prototype to control the generation process, all making generative machine learning more applicable to a real-world scenario of designing a building.
data injection attacks (FDIAs) pose a significant threat to smart power grids. Recent efforts have focused on developing machine learning (ML)-based defense strategies against such attacks. However, existing strategie...
详细信息
data injection attacks (FDIAs) pose a significant threat to smart power grids. Recent efforts have focused on developing machine learning (ML)-based defense strategies against such attacks. However, existing strategies offer limited detection performance since they (a) lack the capability of embedding the spatial aspects of the power system topology in the detection mechanism, (b) offer topology-specific detection that does not generalize well to practical systems with seasonal reconfigurations in their topology, or (c) offer detection based on only seen types of FDIAs present in the training set. Therefore, in this paper, we aim to develop a defense strategy that offers an improved generalization ability and detection performance against unseen attacks. Towards this objective, we propose a graph autoencoder (GAE)-based detection strategy that (a) captures spatio-temporal features of power systems, hence, offering improved detection performance, (b) is trained on comprehensive graphs reflecting various realizations of power system topologies, hence, offering better generalization abilities, and (c) works effectively against unseen FDIAs. The proposed detector is trained and tested on various topological configurations from 14, 39, and 118-bus systems offering detection rates (DRs) of 93.6%, 95.7%, and 99.1%, respectively, when tested against unseen FDIAs and unseen topologies. This presents an improvement of 11.5 - 30% compared to existing ML-based strategies.
Background The appearance of cancer subtypes with different clinical significance fully reflects the high heterogeneity of cancer. At present, the method of multi-omics integration has become more and more mature. How...
详细信息
Background The appearance of cancer subtypes with different clinical significance fully reflects the high heterogeneity of cancer. At present, the method of multi-omics integration has become more and more mature. However, in the practical application of the method, the omics of some samples are *** The purpose of this study is to establish a depth model that can effectively integrate and express partial multi-omics data to accurately identify cancer *** We proposed a novel partial multi-omics learning model for cancer subtypes, MPGIL (Multi-channel Partial graph Integration Learning). MPGIL has two main components. Firstly, it obtains more lateral adjacency information between samples within the omics through the multi-channel graph autoencoders based on high-order proximity. To reduce the negative impact of missing samples, the weighted fusion layer is introduced to replace the concatenate layer to learn the consensus representation across multi-omics. Secondly, a classifier is introduced to ensure that the consensus representation is representative of clustering. Finally, subtypes were identified by *** This study compared MPGIL with other multi-omics integration methods on 16 datasets. The clinical and survival results show that MPGIL can effectively identify subtypes. Three ablation experiments are designed to highlight the importance of each component in MPGIL. A case study of AML was conducted. The differentially expressed gene profiles among its subtypes fully reveal the high heterogeneity of *** MPGIL can effectively learn the consistent expression of partial multi-omics datasets and discover subtypes, and shows more significant performance than the state-of-the-art methods.
How can we exploit Label Propagation (LP) to improve the performance of GNN models on heterophilic graphs? graph Neural Network (GNN) models have received a lot of attention as a powerful deep learning technology that...
详细信息
ISBN:
(纸本)9798350370027;9798350370034
How can we exploit Label Propagation (LP) to improve the performance of GNN models on heterophilic graphs? graph Neural Network (GNN) models have received a lot of attention as a powerful deep learning technology that uses graph structure and features, and has achieved an archived state-of-the-art performance for graph-related tasks. LP has been applied in various studies to improve performance of GNN models. However, LP does not perform well on heterophilic graphs, where nodes of different types are linked with each other, since LP assumes that the graphs inherently exhibits homophily, where similar nodes tend to be linked. Such heterophilic graphs are increasingly common nowadays. In this paper, we propose LPkG (Label Propagation on kNearest Neighbor graphs of graph autoencoder), a simple but effective method to engage LP to improve the performance of GNN models even on heterophilic graphs. LPkG constructs a supplementary homophilic graph, peforms LP on this graph, and uses the results together with the results of GNN models. The supplementary graph is a k-Nearest Neighbor (k-NN) graph genereated from a latent space computed by graph autoencoder (GAE). Experimental results demonstrate that LPkG consistently achieves performance improvement on various heterophilic graph datasets: 2.75% on the Wisconsin dataset, 2.23% on the Texas dataset, and 2.55% on the Cornell dataset.
graph Convolutional Neural Networks (GCN) is a rapidly advancing deep learning algorithm for learning graph representations. One limitation of GCN is that it cannot guarantee optimal low-pass characteristics, thus str...
详细信息
ISBN:
(纸本)9798350370058;9798350370164
graph Convolutional Neural Networks (GCN) is a rapidly advancing deep learning algorithm for learning graph representations. One limitation of GCN is that it cannot guarantee optimal low-pass characteristics, thus struggling to effectively filter out high-frequency noises. In this study, we propose a GCN -based autoencoder that addresses this issue by incorporating Laplace-based filters for high-frequency noise reduction. To address potential overfitting caused by the mean-square error loss function used for reconstructing feature matrices, which is sensitive to the number of vector paradigms and dimensions, we introduce a cosine error loss function to mitigate this impact. Additionally, we employ a feature enhancement strategy during training. Through experiments conducted on three real-world datasets, we demonstrate that our proposed autoencoder clustering algorithm outperforms baseline graph representation learning algorithms in node clustering tasks. Furthermore, we assess the parameter sensitivity of our algorithm through extensive experiments.
Analyzing the community structure of brain networks provides new insights into human brain function. Existing studies broadly use conventional network clustering approaches. While graph neural networks have recently s...
详细信息
ISBN:
(纸本)9798350349405;9798350349399
Analyzing the community structure of brain networks provides new insights into human brain function. Existing studies broadly use conventional network clustering approaches. While graph neural networks have recently shown promise in modeling brain functional connectivity (FC) networks, their applications to brain community detection still need improvement and further refinement. Moreover, identifying common community structure while resolving the single-subject partitions across multiple individual networks remains underexplored. We propose a Deep Multi-graph Embedded Clustering (DMGEC) framework to identify shared community partition in brain FC networks over a cohort of individuals. By incorporating the consensus information aggregated across network structures, DMGEC leverages a graph autoencoder to produce consensus-aware latent representations of individual networks, and applies deep embedded clustering on the multi-subject network representation to produce common community assignment of brain nodes. Simulations show superior community recovery by our method compared to conventional approaches, especially for networks with large number of communities. When applied to functional magnetic resonance imaging (fMRI) data, the DMGEC achieves outstanding alikeness over individual partitions, and uncovers group-level differences in brain community motifs between major depressive disorder patients and normal controls.
Group recommendation has recently drawn a lot of attention to the recommender system community. Currently, several deep learning-based approaches are leveraged to learn preferences of groups for items and predict next...
详细信息
Group recommendation has recently drawn a lot of attention to the recommender system community. Currently, several deep learning-based approaches are leveraged to learn preferences of groups for items and predict next items in which groups may be interested. Yet, their recommendation performance is still unsatisfactory due to sparse group-item interactions. To address this challenge, this study presents a novel model, called group recommendation model with two-stage deep learning (GRMTDL), which encompasses two sequential stages: 1) group representation learning (GRL) and 2) group preference learning (GPL). In GRL, we first construct an undirected tripartite graph over group-user-item interactions, and then employ it to accurately learn group semantic features through a spatial-based variational graph autoencoder network. While in GPL, we first introduce a dual PL-network that contains two structure-sharing subnetworks: 1) group PL-network employed for GPL and 2) user PL-network utilized for user preference learning. Then, we design a novel layered transfer learning (LTL) method to learn group preferences by alternately optimizing these two subnetworks. In particular, it can effectively absorb knowledge of user preferences into the process of GPL. Furthermore, extensive experiments on four real-world datasets demonstrate that the proposed GRMTDL model outperforms the state-of-the-art baselines for group recommendation.
暂无评论