Latent Graph Research Articles

On a wider class of prior distributions for graphical models

AbstractGaussian graphical models are useful tools for conditional independence structure inference of multivariate random variables. Unfortunately, Bayesian inference of latent graph structures is challenging due to exponential growth of $\mathcal{G}_n$ , the set of all graphs in n vertices. One approach that has been proposed to tackle this problem is to limit search to subsets of $\mathcal{G}_n$ . In this paper we study subsets that are vector subspaces with the cycle space $\mathcal{C}_n$ as the main example. We propose a novel prior on $\mathcal{C}_n$ based on linear combinations of cycle basis elements and present its theoretical properties. Using this prior, we implement a Markov chain Monte Carlo algorithm, and show that (i) posterior edge inclusion estimates computed with our technique are comparable to estimates from the standard technique despite searching a smaller graph space, and (ii) the vector space perspective enables straightforward implementation of MCMC algorithms.

Lyapunov exponents for temporal networks.

By interpreting a temporal network as a trajectory of a latent graph dynamical system, we introduce the concept of dynamical instability of a temporal network and construct a measure to estimate the network maximum Lyapunov exponent (nMLE) of a temporal network trajectory. Extending conventional algorithmic methods from nonlinear time-series analysis to networks, we show how to quantify sensitive dependence on initial conditions and estimate the nMLE directly from a single network trajectory. We validate our method for a range of synthetic generative network models displaying low- and high-dimensional chaos and finally discuss potential applications.

Attention-Based Multiple Graph Convolutional Recurrent Network for Traffic Forecasting

Traffic forecasting is essential in the development of intelligent transportation systems, as it enables the formulation of effective traffic dispatching strategies and contributes to the reduction of traffic congestion. The abundance of research focused on modeling complex spatiotemporal correlations for accurate traffic prediction, however many of these prior works perform feature extraction based solely on prior graph structures, thereby overlooking the latent graph connectivity inherent in the data and degrading a decline in prediction accuracy. In this study, we present a novel Attention-based Multiple Graph Convolutional Recurrent Network (AMGCRN) to capture dynamic and latent spatiotemporal correlations in traffic data. The proposed model comprises two spatial feature extraction modules. Firstly, a dot product attention mechanism is utilized to construct an adaptive graph to extract the similarity of road structure. Secondly, the graph attention mechanism is leveraged to enhance the extraction of local traffic flow features. The outputs of these two spatial feature extraction modules are integrated through a gating mechanism and fed into a Gated Recurrent Unit (GRU) to make spatiotemporal interaction predictions. Experimental results on two real-world traffic datasets demonstrate the superiority of the proposed AMGCRN over state-of-the-art baselines. The results suggest that the proposed model is effective in capturing complex spatiotemporal correlations and achieving about 1% improvements in traffic forecasting.

Adaptive Latent Graph Representation Learning for Image-Text Matching.

Image-text matching is a challenging task due to the modality gap. Many recent methods focus on modeling entity relationships to learn a common embedding space of image and text. However, these methods suffer from distractions of entity relationships such as irrelevant visual regions in an image and noisy textual words in a text. In this paper, we propose an adaptive latent graph representation learning method to reduce the distractions of entity relationships for image-text matching. Specifically, we use an improved graph variational autoencoder to separate the distracting factors and latent factor of relationships and jointly learn latent textual graph representations, latent visual graph representations, and a visual-textual graph embedding space. We also introduce an adaptive cross-attention mechanism to perform feature attending on the latent graph representations across images and texts, thus further narrowing the modality gap to boost the matching performance. Extensive experiments on two public datasets, Flickr30K and COCO, show the effectiveness of our method.

Graphical Assistant Grouped Network Autoregression Model: A Bayesian Nonparametric Recourse

Vector autoregression model is ubiquitous in classical time series data analysis. With the rapid advance of social network sites, time series data over latent graph is becoming increasingly popular. In this article, we develop a novel Bayesian grouped network autoregression model, which can simultaneously estimate group information (number of groups and group configurations) and group-wise parameters. Specifically, a graphically assisted Chinese restaurant process is incorporated under the framework of the network autoregression model to improve the statistical inference performance. An efficient Markov chain Monte Carlo sampling algorithm is used to sample from the posterior distribution. Extensive studies are conducted to evaluate the finite sample performance of our proposed methodology. Additionally, we analyze two real datasets as illustrations of the effectiveness of our approach.

Distantly supervised relation extraction with KB-enhanced reconstructed latent iterative graph networks

We investigate distantly supervised relation extraction with knowledge-guided latent graphs and an iterative graph learner. For the relation extraction tasks, we assume that the input sentences contain latent graphs with useful structural information between mentions and relations of the distantly supervised data. We first embed input sentences with default initialized graphs to utilize this information. We also used the pre-trained Knowledge Bases (KB) to guide the latent space of a Variational Graph Auto-Encoders (VGAE) module. Then the VGAE applies the re-parameterization mechanism to reconstruct the initial graph, injecting extra knowledge into the initial latent graph and reducing its unsteadiness. Subsequently, we optimize the latent graph structures and their corresponding node embeddings simultaneously by the iterative graph learner, and the better latent graphs can improve the downstream relation extraction task. Experiment results on three datasets (NYT10, WIKIDISTANT and GIDS) show that latent graph learning helps our model perform better than previous works.

Correlations of network trajectories

Temporal networks model how the interaction between elements in a complex system evolve over time. Just like complex systems display collective dynamics, here we interpret temporal networks as trajectories performing a collective motion in graph space, following a latent graph dynamical system. Under this paradigm, we propose a way to measure how the network pulsates and collectively fluctuates over time and space. To this aim, we extend the notion of linear correlations function to the case of sequences of network snapshots, i.e. a network trajectory. We construct stochastic and deterministic graph dynamical systems and show that the emergent collective correlations are well captured by simple measures, and illustrate how these patterns are revealed in empirical networks arising in different domains.

Multi-Modal Graph Learning for Disease Prediction.

Benefiting from the powerful expressive capability of graphs, graph-based approaches have been popularly applied to handle multi-modal medical data and achieved impressive performance in various biomedical applications. For disease prediction tasks, most existing graph-based methods tend to define the graph manually based on specified modality (e.g., demographic information), and then integrated other modalities to obtain the patient representation by Graph Representation Learning (GRL). However, constructing an appropriate graph in advance is not a simple matter for these methods. Meanwhile, the complex correlation between modalities is ignored. These factors inevitably yield the inadequacy of providing sufficient information about the patient's condition for a reliable diagnosis. To this end, we propose an end-to-end Multi-modal Graph Learning framework (MMGL) for disease prediction with multi-modality. To effectively exploit the rich information across multi-modality associated with the disease, modality-aware representation learning is proposed to aggregate the features of each modality by leveraging the correlation and complementarity between the modalities. Furthermore, instead of defining the graph manually, the latent graph structure is captured through an effective way of adaptive graph learning. It could be jointly optimized with the prediction model, thus revealing the intrinsic connections among samples. Our model is also applicable to the scenario of inductive learning for those unseen data. An extensive group of experiments on two disease prediction tasks demonstrates that the proposed MMGL achieves more favorable performance. The code of MMGL is available at https://github.com/SsGood/MMGL.

Sprod for de-noising spatially resolved transcriptomics data based on position and image information.

Spatially resolved transcriptomics (SRT) provide gene expression close to, or even superior to, single-cell resolution while retaining the physical locations of sequencing and often also providing matched pathology images. However, SRT expression data suffer from high noise levels, due to the shallow coverage in each sequencing unit and the extra experimental steps required to preserve the locations of sequencing. Fortunately, such noise can be removed by leveraging information from the physical locations of sequencing, and the tissue organization reflected in corresponding pathology images. In this work, we developed Sprod, based on latent graph learning of matched location and imaging data, to impute accurate SRT gene expression. We validated Sprod comprehensively and demonstrated its advantages over previous methods for removing drop-outs in single-cell RNA-sequencing data. We showed that, after imputation by Sprod, differential expression analyses, pathway enrichment and cell-to-cell interaction inferences are more accurate. Overall, we envision de-noising by Sprod to become a key first step towards empowering SRT technologies for biomedical discoveries.

Latent graph learning with dual-channel attention for relation extraction

As a building block of information retrieval, relation extraction aims at predicting the relation type between two given entities in a piece of text. This task becomes challenging when it is confronted with long text that contains many task-unrelated tokens. Recent attempts to solve this problem have resorted to learning the relatedness among tokens. However, how to obtain appropriate graph for better relatedness representation still remains outstanding, while existing methods have room to improve. In this paper, we propose a novel latent graph learning method to enhance the expressivity of contextual information for the entities of interest. In particular, we design a dual-channel attention mechanism for multi-view graph learning and pool the learned multi-views to sift unrelated tokens for latent graph. This process can be repeated many times for refining the latent structure. We show that our method achieves superior performance on several benchmark datasets, compared to strong baseline models and prior multi-view graph learning approach.

Abstract 1208: Denoising spatial expression profiling data based on in situ position and image information

Abstract Spatial expression profiling (SEP) technologies provide gene expression close to or even superior to single-cell resolution while retaining the physical locations of sequencing and often also providing matched pathology images. However, the expression data captured by SEP technologies suffer from high noise levels, including but not limited to dropouts as in regular single-cell RNA-sequencing (scRNA-seq). The extra experimental steps for preserving the spatial locations of sequencing could result in even more severe noises, compared to regular scRNA-seq. Fortunately, such noises could be largely reduced by leveraging information from the physical locations of sequencing and the tissue and cellular organization reflected by corresponding pathology images. In this work, we demonstrated the extensive levels of noise in SEP data. We developed a mathematical model, named Sprod, to reduce such noises based on latent space and graph learning of matched location and imaging data. We comprehensively validated Sprod and demonstrated its advantages over prior methods for removing dropouts in scRNA-seq data. We further showed that, after adequately de-noising by Sprod, differential expression analyses, pseudotime analyses, and cell-to-cell interaction inferences yield significantly more informative results in various biological application settings. In particular, with Sprod, we discovered 3-4 times more RNA transcripts that were actively transported in mouse hippocampus neurons. We also showed that the tumor cells at the tumor-stroma boundaries demonstrate differential transcriptomic features from the tumor cells in the central regions, caused by their interactions with the stroma/immune cells. Overall, we envision denoising by Sprod to become a key first step to empower SEP technologies for biomedical discoveries and innovations. Citation Format: Yunguan Wang, Bing Song, Shidan Wang, Mengyi Chen, Yang Xie, Guanghua Xiao, Li Wang, Tao Wang. Denoising spatial expression profiling data based on in situ position and image information [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1208.

AvgNet: Adaptive Visibility Graph Neural Network and Its Application in Modulation Classification

Our digital world is full of time series and graphs which capture the various aspects of many complex systems. Traditionally, there are respective methods in processing these two different types of data, e.g., Recurrent Neural Network (RNN) and Graph Neural Network (GNN), while in recent years, time series could be mapped to graphs by using the techniques such as Visibility Graph (VG), which simultaneously captures relevant aspects of both local and global dynamics in an easy way, so that researchers can use graph algorithms to mine the knowledge in time series and gain special latent graph representation features. Such mapping methods establish a bridge between time series and graphs, and have high potential to facilitate the analysis of various real-world time series. However, the VG method and its variants are just based on fixed rules and thus lack of flexibility, largely limiting their application in reality. In this paper, we propose an Adaptive Visibility Graph (AVG) algorithm that can adaptively map time series into graphs, based on which we further establish an end-to-end classification framework AvgNet, by utilizing GNN model DiffPool as the classifier. We then adopt AvgNet for radio signal modulation classification which is an important task in the field of wireless communication. The simulations validate that AvgNet outperforms a series of advanced deep learning methods, achieving the state-of-the-art performance in this task.

Learning ordinal constraint binary codes for fast similarity search

Similarity search with hashing has become one of the fundamental research topics in computer vision and multimedia. The current researches on semantic-preserving hashing mainly focus on exploring the semantic similarities between pointwise or pairwise samples in the visual space to generate discriminative hash codes. However, such learning schemes fail to explore the intrinsic latent features embedded in the high-dimensional feature space and they are difficult to capture the underlying topological structure of data, yielding low-quality hash codes for image retrieval. In this paper, we propose an ordinal-preserving latent graph hashing (OLGH) method, which derives the objective hash codes from the latent space and preserves the high-order locally topological structure of data into the learned hash codes. Specifically, we conceive a triplet constrained topology-preserving loss to uncover the ordinal-inferred local features in binary representation learning. By virtue of this, the learning system can implicitly capture the high-order similarities among samples during the feature learning process. Moreover, the well-designed latent subspace learning is built to acquire the noise-free latent features based on the sparse constrained supervised learning. As such, the latent under-explored characteristics of data are fully employed in subspace construction. Furthermore, the latent ordinal graph hashing is formulated by jointly exploiting latent space construction and ordinal graph learning. An efficient optimization algorithm is developed to solve the resulting problem to achieve the optimal solution. Extensive experiments conducted on diverse datasets show the effectiveness and superiority of the proposed method when compared to some advanced learning to hash algorithms for fast image retrieval. The source codes of this paper are available at https://github.com/DarrenZZhang/OLGH .

Coastal wetland hyperspectral classification under the collaborative of subspace partition and infinite probabilistic latent graph ranking

Coastal wetland hyperspectral classification under the collaborative of subspace partition and infinite probabilistic latent graph ranking

Powerful multiple testing of paired null hypotheses using a latent graph model

In this paper, we explore the multiple testing problem of paired null hypotheses, for which the data are collected on pairs of entities and tests have to be performed for each pair. Typically, for each pair (i,j), we observe some interaction/association score between i and j and the aim is to detect the pairs with a significant score. In this setting, it is natural to assume that the true/false null constellation is structured according to an unobserved graph, where present edges correspond to a significant association score. The point of this work is to build an improved multiple testing decision by learning the graph structure. Our approach is in line with the seminal work of Sun and Cai [46], that uses the hidden Markov model to structure the dependencies between null hypotheses. Here, we adapt this strategy by considering the stochastic block model for the latent graph. Under appropriate assumptions, the new proposed procedure is shown to control the false discovery rate, up to remainder terms that vanish when the size of the number of hypotheses increases. The procedure is also shown to be nearly optimal in the sense that it is close to the procedure maximizing the true discovery rate. Numerical experiments reveal that our method outperforms state-of-the-art methods and is robust to model mis-specification. Finally, the applicability of the new method is demonstrated on data concerning the usage of self-service bicycles in London.

Graph-based knowledge tracing: Modeling student proficiency using graph neural networks

Recent advancements in computer-assisted learning systems have caused an increase in the research in knowledge tracing, wherein student performance is predicted over time. Student coursework can potentially be structured as a graph. Incorporating this graph-structured nature into a knowledge tracing model as a relational inductive bias can improve its performance; however, previous methods, such as deep knowledge tracing, did not consider such a latent graph structure. Inspired by the recent successes of graph neural networks (GNNs), we herein propose a GNN-based knowledge tracing method, i.e., graph-based knowledge tracing. Casting the knowledge structure as a graph enabled us to reformulate the knowledge tracing task as a time-series node-level classification problem in the GNN. As the knowledge graph structure is not explicitly provided in most cases, we propose various implementations of the graph structure. Empirical validations on two open datasets indicated that our method could potentially improve the prediction of student performance and demonstrated more interpretable predictions compared to those of the previous methods, without the requirement of any additional information.

Multi-label transfer learning via latent graph alignment

Multi-label transfer learning via latent graph alignment

Multimodal Earth observation data fusion: Graph-based approach in shared latent space

Multiple and heterogenous Earth observation (EO) platforms are broadly used for a wide array of applications, and the integration of these diverse modalities facilitates better extraction of information than using them individually. The detection capability of the multispectral unmanned aerial vehicle (UAV) and satellite imagery can be significantly improved by fusing with ground hyperspectral data. However, variability in spatial and spectral resolution can affect the efficiency of such dataset's fusion. In this study, to address the modality bias, the input data was projected to a shared latent space using cross-modal generative approaches or guided unsupervised transformation. The proposed adversarial networks and variational encoder-based strategies used bi-directional transformations to model the cross-domain correlation without using cross-domain correspondence. It may be noted that an interpolation-based convolution was adopted instead of the normal convolution for learning the features of the point spectral data (ground spectra). The proposed generative adversarial network-based approach employed dynamic time wrapping based layers along with a cyclic consistency constraint to use the minimal number of unlabeled samples, having cross-domain correlation, to compute a cross-modal generative latent space. The proposed variational encoder-based transformation also addressed the cross-modal resolution differences and limited availability of cross-domain samples by using a mixture of expert-based strategy, cross-domain constraints, and adversarial learning. In addition, the latent space was modelled to be composed of modality independent and modality dependent spaces, thereby further reducing the requirement of training samples and addressing the cross-modality biases. An unsupervised covariance guided transformation was also proposed to transform the labelled samples without using cross-domain correlation prior. The proposed latent space transformation approaches resolved the requirement of cross-domain samples which has been a critical issue with the fusion of multi-modal Earth observation data. This study also proposed a latent graph generation and graph convolutional approach to predict the labels resolving the domain discrepancy and cross-modality biases. Based on the experiments over different standard benchmark airborne datasets and real-world UAV datasets, the developed approaches outperformed the prominent hyperspectral panchromatic sharpening, image fusion, and domain adaptation approaches. By using specific constraints and regularizations, the network developed was less sensitive to network parameters, unlike in similar implementations. The proposed approach illustrated improved generalizability in comparison with the prominent existing approaches. In addition to the fusion-based classification of the multispectral and hyperspectral datasets, the proposed approach was extended to the classification of hyperspectral airborne datasets where the latent graph generation and convolution were employed to resolve the domain bias with a small number of training samples. Overall, the developed transformations and architectures will be useful for the semantic interpretation and analysis of multimodal data and are applicable to signal processing, manifold learning, video analysis, data mining, and time series analysis, to name a few.

Neural Relational Inference with Efficient Message Passing Mechanisms

Many complex processes can be viewed as dynamical systems of interacting agents. In many cases, only the state sequences of individual agents are observed, while the interacting relations and the dynamical rules are unknown. The neural relational inference (NRI) model adopts graph neural networks that pass messages over a latent graph to jointly learn the relations and the dynamics based on the observed data. However, NRI infers the relations independently and suffers from error accumulation in multi-step prediction at dynamics learning procedure. Besides, relation reconstruction without prior knowledge becomes more difficult in more complex systems. This paper introduces efficient message passing mechanisms to the graph neural networks with structural prior knowledge to address these problems. A relation interaction mechanism is proposed to capture the coexistence of all relations, and a spatio-temporal message passing mechanism is proposed to use historical information to alleviate error accumulation. Additionally, the structural prior knowledge, symmetry as a special case, is introduced for better relation prediction in more complex systems. The experimental results on simulated physics systems show that the proposed method outperforms existing state-of-the-art methods.

Multi-scale skeleton adaptive weighted GCN for skeleton-based human action recognition in IoT

Skeleton-based human action recognition has become a hot topic due to its potential advantages. Graph convolution network (GCN) has obtained remarkable performances in the modeling of skeleton-based human action recognition in IoT. In order to capture robust spatial–temporal features from the human skeleton, a powerful feature extractor is essential. However, Most GCN-based methods use the fixed graph topology. Besides, only a single-scale feature is used, and the multi-scale information is ignored. In this paper, we propose a multi-scale skeleton adaptive weighted graph convolution network (MS-AWGCN) for skeleton-based action recognition. Specifically, a multi-scale skeleton graph convolution network is adopted to extract more abundant spatial features of skeletons. Moreover, we develop a simple graph vertex fusion strategy, which can learn the latent graph topology adaptively by replacing the handcrafted adjacency matrix with a learnable matrix. According to different sampling strategies, weighted learning method is adopted to enrich features while aggregating. Experiments on three large datasets illustrate that the proposed method achieves comparable performances to state-of-the-art methods. Our proposed method attains an improvement of 0.9% and 0.7% respectively over the recent GCN-based method on the NTU RGB+D and Kinetics dataset.