Gaussian Graph Research Articles

Contrastive cross-domain sequential recommendation via emphasized intention features

The objective of cross-domain sequential recommendation is to forecast upcoming interactions by leveraging past interactions across diverse domains. Most methods aim to utilize single-domain and cross-domain information as much as possible for personalized preference extraction and effective integration. However, on one hand, most models ignore that cross-domain information is composed of multiple single-domains when generating representations. They still treat cross-domain information the same way as single-domain information, resulting in noisy representation generation. Only by imposing certain constraints on cross-domain information during representation generation can subsequent models minimize interference when considering user preferences. On the other hand, some methods neglect the joint consideration of users’ long-term and short-term preferences and reduce the weight of cross-domain user preferences to minimize noise interference. To better consider the mutual promotion of cross-domain and single-domains factors, we propose a novel model (C2DREIF) that utilizes Gaussian graph encoders to handle information, effectively constraining the correlation of information and capturing useful contextual information more accurately. It also employs a Top-down transformer to accurately extract user intents within each domain, taking into account the user’s long-term and short-term preferences. Additionally, entropy regularized is applied to enhance contrastive learning and mitigate the impact of randomness caused by negative sample composition.

Structured discriminative Gaussian graph learning for multimode process monitoring

AbstractAiming at the actual industrial process background that different modes share the same system configurations and control structure, this article proposes a novel structured discriminant Gaussian graph learning for multimode process monitoring. The proposed method considers not only the sparsity of graph model but also the measurement of data variation based on a mismatched graph and the common node support between different graphical structures. The objective function involves two sets of regularization terms: the trace terms for mismatched measurements and the ‐norm imposed on the union of decomposed graph matrices. Due to the introduced mismatched trace terms, the cost of matching the data points and graph models that have inconsistent class labels can be expanded, which brings more discrimination for the graph‐based mode identification. While the common structure extracted by the ‐norm forces the estimated graph models to have structural similarities, thus alleviating the negative influence caused by graph discrimination. Once a relatively accurate and discriminative reference graph model is obtained, the downstream test graph learning and analysis can be conducted online by employing the moving window techniques. By comparing the matched and mismatched graph‐based measurements, the process mode can be identified correctly and stably. To grasp the abnormal process changes, the ‐norm for the row sparsity is again applied to the graph difference matrices, the sensitive monitoring statistics and the fault isolation results can be obtained effectively. All the optimization problems in this paper can be solved using the alternating direction multiplier (ADMM) algorithm. The effectiveness of our proposed approach is illustrated by the application to a real blast furnace iron‐making production process.

REMI: Few-Shot ISAR Target Classification via Robust Embedding and Manifold Inference.

Unknown image deformation and few-shot issues have posed significant challenges to inverse synthetic aperture radar (ISAR) target classification. To achieve robust feature representation and precise correlation modeling, this article proposes a novel two-stage few-shot ISAR classification network, dubbed as robust embedding and manifold inference (REMI). In the robust embedding stage, a multihead spatial transformation network (MH-STN) is designed to adjust unknown image deformations from multiple perspectives. Then, the grouped embedding network (GEN) integrates and compresses diverse information by grouped feature extraction, intermediate feature fusion, and global feature embedding. In the manifold inference stage, a masked Gaussian graph attention network (MG-GAT) is devised to capture the irregular manifold of samples in the embedding space. In particular, the node features are described by Gaussian distributions, with interactions guided by the masked attention mechanism. Experimental results on two ISAR datasets demonstrate that REMI significantly improves the performance of few-shot classification and exhibits robustness in various scenarios.

Risk assessment of computer network information using a proposed approach: Fuzzy hierarchical reasoning model based on scientific inversion parallel programming

Abstract Background Computer networks are involved in many fields such as business, education, marketing, government, and tourism in several forms. Technologies related to security protection and improvement of information integrity are used and developed for computer networks intruded on by unauthorized people and help save their confidentiality. Methods To improve the risk identification of computer networks, this manuscript combined a fuzzy hierarchical reasoning model with the scientific inversion parallel programming method to study the risk of computer networks. Moreover, this article defined and analyzed a d-order neighborhood message propagation algorithm. A d-order neighborhood parallel message propagation algorithm using the Gaussian graph model was proposed. Results The risk of the computer network was analyzed using the proposed method resulting in better protection effectiveness. Conclusion The simulations showed that the proposed algorithm could effectively detect risks and improve the security of the computer network.

Exploring food intake networks and anhedonia symptoms in a Chilean Adults sample

Introductionhigh-calorie and sugar-sweetened food is considered more pleasant food. People with anhedonia symptoms have difficulties experiencing pleasure in daily activities. However, is still unclear if anhedonia symptomatology increases palatable food consumption in the Chilean Adults sample. Objectiveto explore food networks in the Chilean Adults sample and in people with anhedonia symptom. Methodsthe sample was recruited through digital platforms. Pregnant or lactating women and subjects under pharmacological treatment or psychotherapy were excluded. A total of 1242 subjects, 76.6% women, with a mean age of 30.7 (SD 9.3) and who were highly educated, participated in the study. Data were collected through an online survey. A questionnaire on food consumption based on daily and weekly frequency was used, as well as the Snaith-Hamilton Pleasure Scale to measure anhedonia symptoms. We employed the Gaussian graph model (GGM) to analyze food consumption as networks. We started with the total sample, and then we repeated the analysis on a subsample with anhedonia symptoms, and next on a subsample with exclusively food-related anhedonia. Resultsin the total sample, a positive and strong relationship was observed between fruits and vegetables, as well as a negative association with the triad of sugar-sweetened beverages, fast food, and fried food. The network in anhedonic subjects shows that “pasta, rice & potatoes” and “bread” have a stronger association and a more central place in the network compared those without anhedonia symptoms. ConclusionsSubjects with anhedonia symptoms have a more central consumption of foods with a high or medium glycemic index compared to subjects without anhedonia symptoms, which could trigger the development of chronic diet-related diseases.

Unveiling common psychological characteristics of proneness to aggression and general psychopathology in a large community youth cohort

Elevated aggression in individuals with psychiatric disorders is frequently reported yet aggressive acts among people with mental illness are often intertwined with proneness to aggression and other risk factors. Evidence has suggested that both general psychopathology and proneness to aggression may share common psychological characteristics. This study aims to investigate the complex relationship between general psychopathology, proneness to aggression, and their contributing factors in community youth. Here, we first examined the association between proneness to aggression and the level of general psychopathology in 2184 community youths (male: 41.2%). To identify common characteristics, we trained machine learning models using LASSO based on 230 features covering sociodemographic, cognitive functions, lifestyle, well-being, and psychological characteristics to predict levels of general psychopathology and proneness to aggression. A subsequent Gaussian Graph Model (GGM) was fitted to understand the relationships between the general psychopathology, proneness to aggression, and selected features. We showed that proneness to aggression was associated with a higher level of general psychopathology (discovery: r = 0.56, 95% CI: [0.52–0.59]; holdout: r = 0.60, 95% CI: [0.54–0.65]). The LASSO model trained on the discovery dataset for general psychopathology was able to predict proneness to aggression in the holdout dataset with a moderate correlation coefficient of 0.606. Similarly, the model trained on the proneness to aggression in the discovery dataset was able to predict general psychopathology in the holdout dataset with a correlation coefficient of 0.717. These results suggest that there is substantial shared information between the two outcomes. The GGM model revealed that isolation and impulsivity factors were directly associated with both general psychopathology and proneness to aggression. These results revealed shared psychological characteristics of general psychopathology and proneness to aggression in a community sample of youths.

Exact test theory in Gaussian graphical models

In this paper, we derive several statistical tests on the precision matrix with application to the determination of the structure of an undirected Gaussian graph. The exact distributions of the test statistics are obtained under the null hypotheses, while the exact distributions of the random matrices, which are used in the construction of the test statistics, are deduced under the alternative hypothesis. Moreover, we present the high-dimensional asymptotic distributions of the test statistics under the null hypothesis. The testing problems that an undirected Gaussian graph possesses a structure that corresponds to the precision matrix of an AR(1) process, to the block-diagonal precision matrix and to the precision of a factor model are discussed in detail. The performance of the proposed statistical tests is further investigated via an extensive simulation study and compared to the benchmark approach.

Anomaly Detection of Servomotors Subject to Highly Accelerated Limit Testing

Companies utilize highly accelerated limit testing (HALT) to ensure efficient product development by accelerating loading conditions in the qualification process. The aim of qualitative accelerated testing such as HALT is to properly identify early behavioral anomalies. To this end, this study utilizes machine learning techniques for detecting anomalies in servomotors in electronic products subjected to HALT. A case study was conducted using a programmable robot kit with 12 servomotors. HALT comprises five types of stress: thermal conditioning (cold and heat), rapid thermal change, vibration, and combined stresses. The anomalous behavior of a servomotor can be identified using a k-nearest neighbor algorithm and verified by inspection using the loading conditions and electrical responses. In addition, anomalous behaviors among servomotors and a control board are assessed using a Gaussian graph model approach. Changes in the Gaussian graph are assessed as anomaly scores using Kullback–Leibler divergence. The anomaly score increased earlier than the operating limit defined by inspection, i.e., the deviation from the initial position of the shaft. The machine learning algorithm successfully identified the failure precursor of the unit. The proposed approach of HALT with the machine learning algorithm supports prognostic health management of servomotors.

Extraction of Joint Entity and Relationships with Soft Pruning and GlobalPointer

In recent years, scholars have paid increasing attention to the joint entity and relation extraction. However, the most difficult aspect of joint extraction is extracting overlapping triples. To address this problem, we propose a joint extraction model based on Soft Pruning and GlobalPointer, short for SGNet. In the first place, the BERT pretraining model is used to obtain the text word vector representation with contextual information, and then the local and non-local information of the word vector is obtained through graph operations. Specifically, to address the lack of information caused by the rule-based pruning strategies, we utilize the Gaussian Graph Generator and the attention-guiding layer to construct a fully connected graph. This process is called soft pruning for short. Then, to achieve node message passing and information integration, we employ GCNs and a thick connection layer. Next, we use the GlobalPointer decoder to convert triple extraction into quintuple extraction to tackle the problem of problematic overlapping triples extraction. The GlobalPointer decoder, unlike the typical feedforward neural network (FNN), can perform joint decoding. In the end, to evaluate the model performance, the experiment was carried out on two public datasets: the NYT and WebNLG. The experiments show that SGNet performs substantially better on overlapping extraction and achieves good results on two publicly available datasets.

On Symmetry of Complete Graphs over Quadratic and Cubic Residues

In this study, we investigate two graphs, one of which has units of a ring Z n as vertices (or nodes) and an edge will be built between two vertices u and v if and only if u 3 ≡ v 3 mod n . This graph will be termed as cubic residue graph. While the other is called Gaussian quadratic residue graph whose vertices are the elements of a Gaussian ring Z n i of the form α = a + i b , β = c + i d , where a , b , c , d are the units of Z n . Two vertices α and β are adjacent to each other if and only if α 2 ≡ β 2 mod n . In this piece of work, we characterize cubic and Gaussian quadratic residue graphs for each positive integer n in terms of complete graphs.

GDPNet: Refining Latent Multi-View Graph for Relation Extraction

Relation Extraction (RE) is to predict the relation type of two entities that are mentioned in a piece of text, e.g., a sentence or a dialogue. When the given text is long, it is challenging to identify indicative words for the relation prediction. Recent advances on RE task are from BERT-based sequence modeling and graph-based modeling of relationships among the tokens in the sequence. In this paper, we propose to construct a latent multi-view graph to capture various possible relationships among tokens. We then refine this graph to select important words for relation prediction. Finally, the representation of the refined graph and the BERT-based sequence representation are concatenated for relation extraction. Specifically, in our proposed GDPNet (Gaussian Dynamic Time Warping Pooling Net), we utilize Gaussian Graph Generator (GGG) to generate edges of the multi-view graph. The graph is then refined by Dynamic Time Warping Pooling (DTWPool). On DialogRE and TACRED, we show that GDPNet achieves the best performance on dialogue-level RE, and comparable performance with the state-of-the-arts on sentence-level RE. Our code is available at https://github.com/XueFuzhao/GDPNet.

Indirect Gaussian Graph Learning Beyond Gaussianity

This paper studies how to capture dependency graph structures from real data that may not be multivariate Gaussian. Starting from marginal loss functions not necessarily derived from probability distributions, we utilize an additive over-parametrization with shrinkage to incorporate variable dependencies into the criterion. An iterative Gaussian graph learning algorithm is proposed with ease in implementation. Statistical analysis shows that the estimators achieve satisfactory accuracy with the error measured in terms of a proper Bregman divergence. Real-life examples in different settings are given to demonstrate the efficacy of the proposed methodology.

Convergence rate analysis of Gaussian belief propagation for Markov networks

Gaussian belief propagation algorithm ( GaBP ) is one of the most important distributed algorithms in signal processing and statistical learning involving Markov networks. It is well known that the algorithm correctly computes marginal density functions from a high dimensional joint density function over a Markov network in a finite number of iterations when the underlying Gaussian graph is acyclic. It is also known more recently that the algorithm produces correct marginal means asymptotically for cyclic Gaussian graphs under the condition of walk summability ( or generalised diagonal dominance ) . This paper extends this convergence result further by showing that the convergence is exponential under the generalised diagonal dominance condition, and provides a simple bound for the convergence rate. Our results are derived by combining the known walk summability approach for asymptotic convergence analysis with the control systems approach for stability analysis.

Bayesian inference for latent chain graphs

In this article we consider Bayesian inference for partially observed Andersson-Madigan-Perlman (AMP) Gaussian chain graph (CG) models. Such models are of particular interest in applications such as biological networks and financial time series. The model itself features a variety of constraints which make both prior modeling and computational inference challenging. We develop a framework for the aforementioned challenges, using a sequential Monte Carlo (SMC) method for statistical inference. Our approach is illustrated on both simulated data as well as real case studies from university graduation rates and a pharmacokinetics study.

Graph Laplacian Regularization for Robust Optical Flow Estimation.

This paper proposes graph Laplacian regularization for robust estimation of optical flow. First, we analyze the spectral properties of dense graph Laplacians and show that dense graphs achieve a better trade-off between preserving flow discontinuities and filtering noise, compared with the usual Laplacian. Using this analysis, we then propose a robust optical flow estimation method based on Gaussian graph Laplacians. We revisit the framework of iteratively reweighted least-squares from the perspective of graph edge reweighting, and employ the Welsch loss function to preserve flow discontinuities and handle occlusions. Our experiments using the Middlebury and MPI-Sintel optical flow datasets demonstrate the robustness and the efficiency of our proposed approach.

Structured sequential Gaussian graphical models for monitoring time-varying process

This article introduces a structured and sequential Gaussian graphical approach for monitoring time-varying industrial systems. To address the presence of time-varying trends, the article utilizes a sequential update procedure for Gaussian graphical models that is based on a moving window approach. The implementation of this approach requires that the Gaussian graph models are constructed based on two additional regularization terms. For a time-varying graphical structure, the overall trend can be isolated by the first additional regularization term; while abnormal events can be timely captured using the second regularization term. Unlike conventional graphical networks that rely on other methods for fault detection, this article proposes a unified graphical framework for fault detection and isolation. In addition, the optimization problem of the Gaussian graphical model can be solved using the Alternative Direction Method of Multiplier (ADMM) algorithm, which has a linear convergence rate, making it easily applicable to large-scale systems. The performance of this graph-based technique is demonstrated by a simulation example and the application study to an industrial distillation process.

Properties of Gaussian Beam Propagating in Ring Resonator Sensor

In this paper, we deduce the paraxial analytical expression for the Gaussian beam propagating in the sandwich slab system which contained double negative material based on light transfer matrix and generalized Huygens-Fresnel integral equation; the evolution properties of emerging Gaussian beam contour graph intensity distribution on the receiver plane, the relation between beam spot size and negative refractive index coefficient, and beam side transmission view in slab system changed with three negative refractive index parameters are illustrated through numerical examples. What is more, we propose a ring resonator sensor to measure the concentration of NaCl solution on the basis of above theory, of which the operating principle is deliberatively analyzed, the influence of the concave mirror curvature radius on the emerging beam evolution is acquired, the functional relation between the normalized central intensity of the emerging beam, the beam spot size, and NaCl solution concentration is further developed by fit linear method, and the mathematical statistics results reach high precision and linearity. It is expected that the proposed ring resonator sensor and the corresponding conclusions can be useful for precise optical measurement, especially for food safety inspection and medical services of health care.

Constrained likelihood for reconstructing a directed acyclic Gaussian graph.

Directed acyclic graphs are widely used to describe directional pairwise relations. Such relations are estimated by reconstructing a directed acyclic graph's structure, which is challenging when the ordering of nodes of the graph is unknown. In such a situation, existing methods such as the neighbourhood and search-and-score methods have high estimation errors or computational complexities, especially when a local or sequential approach is used to enumerate edge directions by testing or optimizing a criterion locally, as a local method may break down even for moderately sized graphs. We propose a novel approach to simultaneously identifying all estimable directed edges and model parameters, using constrained maximum likelihood with nonconvex constraints. We develop a constraint reduction method that constructs a set of active constraints from super-exponentially many constraints. This, coupled with an alternating direction method of multipliers and a difference convex method, permits efficient computation for large-graph learning. We show that the proposed method consistently reconstructs identifiable directions of the true graph and achieves the optimal performance in terms of parameter estimation. Numerically, the method compares favourably with competitors. A protein network is analysed to demonstrate that the proposed method can make a difference in identifying the network's structure.

Image recognition with missing-features based on gaussian mixture model and graph constrained nonnegative matrix factorization.

The demand for automatically recognizing medical images for screening, reference and management is growing faster than ever. Missing data phenomenon in medical image applications is common existence, and it could be inevitable. In this paper, we have addressed the problem of recognizing medical images with missing-features via Gaussian mixture model (GMM)-based approach. Since training a GMM by directly using high-dimensional feature vectors will result in instability, we have proposed a novel strategy to train the GMM from the corresponding reduced-dimensional one. The proposed method contains training and test phases. The former contains feature extraction, graph constrained nonnegative matrix factorization (NMF), GMM training, and the alternating expectation conditional maximization (AECM) for extending the reduced-dimensional GMM. In test phase, two methods, marginalizing GMM using Bayesian decision (MGBD) and conditional mean imputation (CMI), are applied to impute missing-features. Posterior probability of test images is calculated to identify objects. Experimental results on three real datasets demonstrate the feasibility and efficiency of the proposed scheme.

Modeling of a Thermal Spraying Process by Gaussian Chain Graphs

The class of chain graph models is considered for the analysis of a data set arising from a thermal spraying process. Different methods for the selection of Gaussian chain graph models are reviewed and applied. Freely available software for the semi-automatic model selection of Gaussian chain graph models is used and compared.