Network Decision Research Articles

Collaborate decision network based on cross-modal attention for social media microblog recognition

Recently, social media has gradually become an important topic in public opinion analysis field. Among social media, Microblog is one of the most important platforms because it is short, convenient, mobile and instantaneous. Social media microblog recognition well reflects the attitudes from an enormous big colony to a specific incident, either positive or negative, which can be used for deriving competitive intelligence, marketing strategies, detecting depression and so on. However, the existing methods usually use only text or image from internet but not take advantages of their complementary information to finalize the recognition, it limits the performance and robustness of the algorithms. In this paper, we present a collaborate decision network (CDN) based on cross-modal attention to exploit the discriminative attributes of multi modalities by data- and knowledge joint driven strategy in depth, and further improve the recognition performance. In addition, we collect and construct a visual-text microblog recognition dataset with 2854 samples to support the subsequent research of related fields. Finally, experimental reuslts on the collected dataset show the effectiveness and superiority of the proposed CDN.

Ship detection method based on attention guidance and multi-sample decision making

Single-stage target detection methods have the characteristics of fast training speed and short detection time. However, its feature pyramid network is difficult to suppress the background and noise information of SAR ship images, and the detection head has prediction errors. To address this problem, this paper proposes a detection model based on attention guidance and multi-sample decision for synthetic aperture radar ship detection. Firstly, an attention guidance network is proposed and added to the highest level of the feature pyramid to suppress background and noise interference, thereby improving the representation ability of features. Secondly, a multi-sample decision network is proposed to participate in the prediction of target position. This network alleviates the impact of prediction errors on detection results by increasing the number of samples output in the regression branch. Finally, a novel maximum likelihood loss function is designed. This loss function constructs a maximum likelihood function using the samples output from the multi-sample decision network, which is used to standardize the training of the decision network and further improve the accuracy of target positioning. Taking the RetinaNet network model as the baseline method, compared with the baseline method and the current advanced target detection methods, this method shows the highest detection accuracy on the ship detection dataset SSDD, with AP reaching 52.8%. Compared with the baseline method, the proposed method improves the AP evaluation index 3.4% ∼ 5.7%, and the training parameter Params only increases by 2.03 M, and the frame rate FPS only decreases 0.5Iter/s.

Bounded Rational Decision Networks with Belief Propagation.

Complex information processing systems that are capable of a wide variety of tasks, such as the human brain, are composed of specialized units that collaborate and communicate with each other. An important property of such information processing networks is locality: there is no single global unit controlling the modules, but information is exchanged locally. Here, we consider a decision-theoretic approach to study networks of bounded rational decision makers that are allowed to specialize and communicate with each other. In contrast to previous work that has focused on feedforward communication between decision-making agents, we consider cyclical information processing paths allowing for back-and-forth communication. We adapt message-passing algorithms to suit this purpose, essentially allowing for local information flow between units and thus enabling circular dependency structures. We provide examples that show how repeated communication can increase performance given that each unit's information processing capability is limited and that decision-making systems with too few or too many connections and feedback loops achieve suboptimal utility.

Age Differences in Consulting Others During Decision Making.

Many choices are made in the context of one's social network, but age differences in consulting others during decision-making remain underexplored. This pre-registered study examined age differences in the composition of decision-related social networks and associated social preferences and motivations. Based on prior findings regarding age differences in information-seeking and socioemotional preferences, we predicted that with advanced age people would consult fewer and more emotionally close social partners, consult them less frequently, rate consultations as more helpful, and prioritize social goals during decision-making. An adult lifespan sample (N = 485; Mage = 51.08; SDage = 19.63, rangeage = 18-96 years) considered hypothetical choices and reported how many and what types of people they would consult and how often they would consult them. They also identified their most important decision consultants and evaluated interactions with those individuals. As predicted, age was associated with consulting fewer people less frequently and this was not explained by having a smaller range of potential consultants in one's network. Contrary to predictions, the proportion of close social partners in decision networks and the perceived helpfulness of consultants did not vary significantly by age. Moreover, older adults were less likely than younger adults to prioritize relationship quality over decision quality. Controlling for potential covariates did not diminish the observed age effects. These findings align with the prior literature on age-related decrements in pre-decisional information seeking but suggest that age-related preferences for close social partners and socioemotional goals do not extend to decision-related consultations.

Memory Propagation-based Target-aware Segmentation Tracker with Adaptive Mask-attention Decision Network

Memory Propagation-based Target-aware Segmentation Tracker with Adaptive Mask-attention Decision Network

The bi-level consensus model with dual social networks for group decision making

The pursuit of consensus within social networks is a burgeoning area of research, pivotal for harmonizing decision-making amidst diverse opinions. However, existing studies often neglect the crucial balance between costs and benefits in optimizing consensus outcomes. Addressing this gap, this paper introduces a novel bi-level consensus optimization model within the framework of the dual social network. This model aims to achieve an equilibrium between minimizing costs and maximizing benefits, crucial for sustainable decision-making processes. The dual social network framework incorporates positive and negative interactions stemming from trust and opinion similarities, delineating nodes into close, distant, and mixed types based on their relational dynamics. Central to the model is a heterogeneous cost function that integrates individual influence and opinion adjustment, accounting comprehensively for moderator tolerance and incentivization mechanisms. To solve this multi-faceted optimization challenge, the paper proposes a solution leveraging a multi-objective particle swarm algorithm. Through simulation experiments conducted across four distinct social network decision-making scenarios, including a case study on capital investment in an epidemic response center, the paper validates the efficacy and practical applicability of the algorithm. The results underscore the model’s capability to achieve balanced consensus outcomes, offering insights into optimizing decision processes within complex social environments.

Application of the Semi-Supervised Learning Approach for Pavement Defect Detection.

Road surface quality is essential for driver comfort and safety, making it crucial to monitor pavement conditions and detect defects in real time. However, the diversity of defects and the complexity of ambient conditions make it challenging to develop an effective and robust classification and detection algorithm. In this study, we adopted a semi-supervised learning approach to train ResNet-18 for image feature retrieval and then classification and detection of pavement defects. The resulting feature embedding vectors from image patches were retrieved, concatenated, and randomly sampled to model a multivariate normal distribution based on the only one-class training pavement image dataset. The calibration pavement image dataset was used to determine the defect score threshold based on the receiver operating characteristic curve, with the Mahalanobis distance employed as a metric to evaluate differences between normal and defect pavement images. Finally, a heatmap derived from the defect score map for the testing dataset was overlaid on the original pavement images to provide insight into the network's decisions and guide measures to improve its performance. The results demonstrate that the model's classification accuracy improved from 0.868 to 0.887 using the expanded and augmented pavement image data based on the analysis of heatmaps.

A deep learning modular ECG approach for cardiologist assisted adjudication of atrial fibrillation and atrial flutter episodes

BackgroundDetection of atrial tachyarrhythmias (ATA) on long-term ECG recordings is a pre-requisite to reduce ATA-related adverse events. However, the burden of editing massive ECG data is not sustainable. Deep Learning (DL) algorithms provide improved performances on resting ECG databases. However, results on long-term Holter recordings are scarce. ObjectiveWe aimed to build and evaluate a DL modular software using ECG features well known to cardiologists with a user interface that allows cardiologists to adjudicate the results and drive a second DL analysis. MethodsUsing a large (n=187 recordings, 249,419 one-minute samples), beat-to-beat annotated, two-lead Holter database, we built a DL algorithm with a modular structure mimicking expert physician ECG interpretation to classify atrial rhythms. The DL network includes 3 modules (cardiac rhythm regularity, electrical atrial waveform, and raw voltage by time data) followed by a decision network and a long-term weighting factor . The algorithm was validated on an external database. ResultsF1 scores of our classifier were 99% for ATA detection, 95% for Atrial fibrillation and 90% for Atrial flutter respectively. Using the external MIT-database, the classifier obtains F1-score of 97% for the NSR class and 96% for the ATA class. Residual errors could be corrected by manual deactivation of one module in 7/15 of the recordings with an accuracy < 90%. ConclusionA DL modular software using ECG features well known to cardiologists provided an excellent overall performance. Clinically significant residual errors were most often related to the classification of the atrial arrhythmia type (fibrillation vs. flutter). The modular structure of the algorithm helped to edit and correct the AI-based first-pass analysis and will provide a basis for explainability.

An Intelligent Bait Delivery Control Method for Flight Vehicle Evasion Based on Reinforcement Learning

During aerial combat, when an aircraft is facing an infrared air-to-air missile strike, infrared baiting technology is an important means of penetration, and the strategy of effective delivery of infrared bait is critical. To address this issue, this study proposes an improved deep deterministic policy gradient (DDPG) algorithm-based intelligent bait-dropping control method. Firstly, by modeling the relative motion between aircraft, bait, and incoming missiles, the Markov decision process of aircraft-bait-missile infrared effect was constructed with visual distance and line of sight angle as states. Then, the DDPG algorithm was improved by means of pre-training and classification sampling. Significantly, the infrared bait-dropping decision network was trained through interaction with the environment and iterative learning, which led to the development of the bait-dropping strategy. Finally, the corresponding environment was transferred to the Nvidia Jetson TX2 embedded platform for comparative testing. The simulation results showed that the convergence speed of this method was 46.3% faster than the traditional DDPG algorithm. More importantly, it was able to generate an effective bait-throwing strategy, enabling the aircraft to successfully evade the attack of the incoming missile. The strategy instruction generation time is only about 2.5 ms, giving it the ability to make online decisions.

HybOff: a Hybrid Offloading approach to improve load balancing in fog environments

Load balancing is crucial in distributed systems like fog computing, where efficiency is paramount. Offloading with different approaches is the key to balancing the load in distributed environments. Static offloading (SoA) falls short in heterogeneous networks, necessitating dynamic offloading to reduce latency in time-sensitive tasks. However, prevalent dynamic offloading (PoA) solutions often come with hidden costs that impact sensitive applications, including decision time, networks congested and distance offloading. This paper introduces the Hybrid Offloading (HybOff) algorithm, which substantially enhances load balancing and resource utilization in fog networks, addressing issues in both static and dynamic approaches while leveraging clustering theory. Its goal is to create an uncomplicated low-cost offloading approach that enhances IoT application performance by eliminating the consequences of hidden costs regardless of network size. Experimental results using the iFogSim simulation tool show that HybOff significantly reduces offloading messages, distance, and decision-offloading consequences. It improves load balancing by 97%, surpassing SoA (64%) and PoA (88%). Additionally, it increases system utilization by an average of 50% and enhances system performance 1.6 times and 1.4 times more than SoA and PoA, respectively. In summary, this paper tries to introduce a new offloading approach in load balancing research in fog environments.

Visualization analysis of urban planning assistant decision network 3D system based on intelligent computing

In recent years, with the continuous development of urban construction and the continuous expansion of urban population, the contradiction between spatial planning and actual urban needs in urban construction has become increasingly acute. However, planning decisions based on traditional urban planning renderings and two-dimensional (2D) planar plans are no longer in line with the actual development needs and specific development status of the city. Urban planning 3D auxiliary systems can improve the accuracy and precision of urban planning decisions. By providing planners with detailed 3D spatial models of the urban environment, potential design defects or conflicts can be identified before urban construction starts. This can reduce costs and risks related to urban development and help create a more livable and sustainable urban environment. This article analyzed and studied urban planning problems based on a three-dimensional (3D) auxiliary system of intelligent computing. In order to examine the effectiveness of intelligent computing 3D auxiliary systems in urban planning and construction, three indicators of development sustainability, transportation network convenience, and urban livability were established for effectiveness evaluation. The final experimental results showed that the average score of the 3D urban planning assistance system based on intelligent computing in urban development sustainability was 9.29 points, transportation convenience score was 92.08 points, and urban livability score was 9.249 points.

Reinforcement learning-based order-dispatching optimization in the ride-sourcing service

Since the emergence of ride-sourcing, the dispatching problem has been a hot research issue in the travel market. Previous studies designed various refined algorithms to improve the dispatching efficiency. However, the heterogeneity of passenger cancellation behavior and driver's work patterns is ignored in the modeling of these studies, which may result in the algorithms failing to entirely capture the randomness in the real world, thereby affecting dispatching efficiency. To fill this research gap, this paper employs a discrete choice model and Gaussian mixture model to model passenger cancellation behavior and extract driver working patterns, respectively. Furthermore, a cooperative deep deterministic policy gradient (DDPG) algorithm is proposed for dispatching. The algorithm constructs decision networks separately for drivers with different working patterns to learn the optimal dispatching decision and designs a centralized neural network to coordinate the dispatching of all vehicles to maximize the global benefit. A price incentive strategy is proposed to improve passenger loyalty by providing incentive prices. Moreover, based on the constructed discrete choice model, a simulation environment is designed to describe the cancellation behavior in the real world and support the training of the algorithm. Experimental results based on real data in Beijing show that the cooperative DDPG algorithm can improve the order response rate by 2% to 6% and platform revenue by 3% to 9% in multiple travel periods while providing a better travel experience to passengers. Additionally, the price incentive strategy also effectively reduces passenger cancellation.

Building a decision-support tool to inform sustainability approaches under complexity: Case study on managing wild ruminants

In wildlife management, differing perspectives among stakeholders generate conflicts about how to achieve disparate sustainability goals that include ecological, economic, and sociocultural dimensions. To mitigate such conflicts, decisions regarding wildlife management must be taken thoughtfully. To our knowledge, there exists no integrative modeling framework to inform these decisions, considering all dimensions of sustainability. We constructed a decision-support tool based on stakeholder workshops and a Bayesian decision network to inform management of wild ruminants in the federal state of Lower Austria. We use collaborative decision analysis to compare resource allocations while accounting for trade-offs among dimensions of sustainability. The tool is designed for application by non-technical users across diverse decision-making contexts with particular sets of wildlife management actions, objectives, and uncertainties. Our tool represents an important step toward developing and evaluating a transparent and replicable approach for mitigating wildlife-based conflicts in Europe and beyond.

A Bayesian decision network–based pre-disaster mitigation model for earthquake-induced cascading events to balance costs and benefits on a limited budget

Cascading disasters induced by earthquakes amplify the severity of the initial impact on environment. Although decisionmakers may face uncertainty, an effective mitigation strategy is critical in environmental management. We propose an earthquake-induced cascading disaster mitigation–Bayesian decision network (ECDM-BDN) model to assess pre-disaster mitigating strategies under limited budgets from the perspective of systematic thinking. This model graphically represents the complex relationship among various variables in a seismic hazard system and resistance system based on disaster system theory. It can predict the triggering of cascading events through probabilistic reasoning and identify the key variable accountable for the range of outputs observed through sensitivity analysis. In addition, cost–benefit analyses are carried out by combining Bayesian decision network utility nodes and dynamic programming to obtain a balance between costs and benefits in the context of limited budgets. An earthquake-induced liquefaction served as a case study to demonstrate the proposed model’s effectiveness. Experimental results indicate that the ECDM-BDN model can balance the costs and effects of each pre-disaster mitigating strategy as well as select the optimal one according to the utility value. The proposed model can perform a “white-box” decision-making process, which is expected to guide earthquake-induced cascading event pre-disaster mitigation in cases of limited budgets.

Climate finance across sub-Saharan Africa: Decision trees and network flows

The structure of climate finance flows from donors (multilateral sources) to recipients (sub-Saharan African countries) was studied. This is the first study to provide a comprehensive network structure of the climate finance flows into Africa, based on the global public climate finance governance system. Network theory and decision tree techniques were employed. The results obtained generally fit the multilateral funding units (MFUs) into two categories: central funding units (CFUs), which simultaneously attend to the climate-related projects of many African countries, and the boundary funding units (BFUs), which cater to only a few countries at once. An isolated region with no BFUs was identified. African countries within this group could be more exposed to climate financial risk as they rely on only the CFUs. In general, with the exception of mitigation REDD (reducing emissions from deforestation and forest degradation) climate finance, a disproportionate distribution of climate themes, with particular reference to adaptation finance, was observed across sub-Saharan Africa. This has real implications for equitable resource allocation of climate funds. The need for African-bred region-wide MFUs is recommended.

Inference and Learning in Dynamic Decision Networks Using Knowledge Compilation

Decision making under uncertainty in dynamic environments is a fundamental AI problem in which agents need to determine which decisions (or actions) to make at each time step to maximise their expected utility. Dynamic decision networks (DDNs) are an extension of dynamic Bayesian networks with decisions and utilities. DDNs can be used to compactly represent Markov decision processes (MDPs). We propose a novel algorithm called mapl-cirup that leverages knowledge compilation techniques developed for (dynamic) Bayesian networks to perform inference and gradient-based learning in DDNs. Specifically, we knowledge-compile the Bellman update present in DDNs into dynamic decision circuits and evaluate them within an (algebraic) model counting framework. In contrast to other exact symbolic MDP approaches, we obtain differentiable circuits that enable gradient-based parameter learning.

Cross-Constrained Progressive Inference for 3D Hand Pose Estimation with Dynamic Observer-Decision-Adjuster Networks

Generalization is very important for pose estimation, especially for 3D pose estimation where small changes in the 2D images could trigger structural changes in the 3D space. To achieve generalization, the system needs to have the capability of detecting estimation errors by double-checking the projection coherence between the 3D and 2D spaces and adapting its network inference process based on this feedback. Current pose estimation is one-time feed-forward and lacks the capability to gather feedback and adapt the inference outcome. To address this problem, we propose to explore the concept of progressive inference where the network learns an observer to continuously detect the prediction error based on constraints matching, as well as an adjuster to refine its inference outcome based on these constraints errors. Within the context of 3D hand pose estimation, we find that this observer-adjuster design is relatively unstable since the observer is operating in the 2D image domain while the adjuster is operating in the 3D domain. To address this issue, we propose to construct two sets of observers-adjusters with complementary constraints from different perspectives. They operate in a dynamic sequential manner controlled by a decision network to progressively improve the 3D pose estimation. We refer to this method as Cross-Constrained Progressive Inference (CCPI). Our extensive experimental results on FreiHAND and HO-3D benchmark datasets demonstrate that the proposed CCPI method is able to significantly improve the generalization capability and performance of 3D hand pose estimation.

Analysis of diabetes based on Tree Decision, Logistic Regression and Artificial Neural Network

The alarming rise in diabetes cases globally, particularly amongst the aging population, underscores a pressing need for innovative, accurate, and early diagnostic tools. Amidst this health crisis, the present study delves into the evaluation of three prominent machine learning in predicting diabetes using the Pima Indians Diabetes Database. A comprehensive analysis, considering metrics such as accuracy, precision, recall, F1 score, and ROC curve, revealed Decision Trees as the most efficacious, exhibiting adeptness in balancing precision and recall. Artificial Neural Networks displayed a remarkable ability to identify true positive cases, making it a notable contender in medical diagnostics. In terms of accuracy, Decision Trees led with a rate of 76.19%, followed by Logistic Regression at 74.03%, and Artificial Neural Networks trailing at 72.64%. The findings suggest Decision Trees' potential utility in offering both a nuanced and comprehensive approach to diabetes prediction. However, the constraints imposed by the dataset, and the inherent limitations and sensitivities of the machine learning models, call for a judicious interpretation of the results. This research illuminates the prospective integration of machine learning in healthcare for enhanced, personalized, and timely diabetes management, paving the way for improved patient outcomes and a substantial reduction in global health disparities attributed to diabetes.

Graph-SeTES: A graph based search task extraction using Siamese network

Search task extraction is a sub-process for query suggestion/reformulation, personalized recommendation, and advertisement in search engines and e-commerce platforms. However, they face both internal and external challenges. Internal challenges include short and misspelled queries and incomplete keywords. External challenges include an unknown number of clusters and a limited number of labeled datasets. Deep-learning models require a large amount of data for training; however, search task datasets are rare and small. To overcome these limitations, we proposed Graph-SeTES, which integrates feature extraction with a decision network that utilizes both distance metrics and decision networks. The existing graph-clustering algorithms use the similarities between search query pairs. Because the Siamese network (SN) finds similarities between two objects, it fits search task extraction. Compared with existing models, SN requires fewer parameters for training due to the shared weights. However, it yields good results even with less labeled data, overcoming the external challenges. We benefit from both distance metrics and narrowing of the linear layers for decision networks. Graph-SeTES was compared with state-of-the-art models, and it outperformed its counterparts. The results were 6% better than those of the best baseline on the CSTE dataset, which maintained this performance difference on the WSMC12 dataset.

BPMB: BayesCNNs with perturbed multi-branch structure for robust facial expression recognition

Wild Facial Expression Recognition (FER) task has been a long-standing challenge due to the various forms of uncertainty exist in expression data. When expression data is fed into a convolutional neural network (CNN), the model's estimated parameters also become uncertain. This uncertainty gives rise to concerns regarding the reliability of the recognition results. To quantify these uncertainties and achieve robust performance in the presence of noisy data, this paper introduces a novel model for Wild Facial Expression Recognition: the Bayesian Convolutional Neural Network with Perturbed Multi-Branch Structure (BPMB). This model aims to address uncertainty issues, enabling the network's decisions to become more deterministic with increasing training accuracy. Specifically, BPMB incorporates variational inference (VI) to introduce a probability distribution for the weights. A variational approximation to the true posterior is derived using Bayes by Backprop, involving two convolution operations: one for classification and another for quantifying uncertainty. Furthermore, an exploration is conducted into a lightweight multi-branch structure that leverages Dropout as a random generator to introduce perturbations during the training process, enhancing the model's robustness while extracting deep features. Extensive experiments validate the superiority of the proposed BPMB algorithm over the majority of existing mainstream algorithms on three widely utilized wild datasets.