Root Node Research Articles

Improved Monte Carlo tree search formulation with multiple root nodes for discrete sizing optimization of truss structures

This article proposes a novel reinforcement learning algorithm using an improved Monte Carlo tree search (IMCTS) formulation for the discrete optimum design of truss structures. IMCTS with multiple root nodes includes the update process, best reward, accelerating technique and terminal condition. The update process means that once a final solution is found, it is used as the initial solution for the next search tree. The best reward is used in the backpropagation step. The accelerating technique is introduced by decreasing the width of the search tree and reducing the maximum number of iterations. The agent is trained to minimize the total structural weight under various constraints until the terminal condition is satisfied. The optimal solution is the minimum value of all solutions found by the search trees. Numerical examples show that the agent can find the optimal solution with low computational cost, stably produces an optimal design, and is suitable for multi-objective structural optimization and large-scale structures.

Tree smoothing: Post-hoc regularization of tree ensembles for interpretable machine learning

Random Forests (RFs) are powerful ensemble learning algorithms that are widely used in various machine learning tasks. However, they tend to overfit noisy or irrelevant features, which can result in decreased generalization performance. Post-hoc regularization techniques aim to solve this problem by modifying the structure of the learned ensemble after training. We propose a novel post-hoc regularization via tree smoothing for classification tasks to leverage the reliable class distributions closer to the root node whilst reducing the impact of more specific and potentially noisy splits deeper in the tree. Our novel approach allows for a form of pruning that does not alter the general structure of the trees, adjusting the influence of nodes based on their proximity to the root node. We evaluated the performance of our method on various machine learning benchmark data sets and on cancer data from The Cancer Genome Atlas (TCGA). Our approach demonstrates competitive performance compared to the state-of-the-art and, in the majority of cases, and outperforms it in most cases in terms of prediction accuracy, generalization, and interpretability.

Exploring Communication Barriers in Bridge-Teams: An Innovative Fuzzy-Bayesian Approach

The bridge teams on merchant vessels have a grave responsibility to guarantee the safe navigation and management of ships in the critical waterways of the world. In addition to maintaining effective communication between external stations (other ships-Vessel Traffic Service), it is crucial to ensure continuous internal collaboration among the bridge team in order to fulfil this important task to the highest standard. Nevertheless, the challenging working conditions and harsh environmental factors may impede the uninterrupted flow of information between bridge teams and disrupt the communication. Communication issues among team members are frequently mentioned as a root cause in maritime accident investigation reports. The aim of this research is to propose a novel model for identifying the factors that may cause to inadequate communication among bridge team members, employing a fuzzy Bayesian network (FBN) approach. As indicated by the findings, attitudinal and behavioural barriers exert a greater influence (43.3%) on communication than language barriers (41.5%), representing the most significant factors affecting communication. Environmental barriers and cultural barriers, on the other hand, have comparatively less impact, at 38.7% and 31.2%, respectively. The sensivity analysis also revealed that the root nodes exhibiting the highest degree of impacts were cultural barriers (31.2%), age differences (20.6%), and workplace issues (20.2%). The findings suggest that bridge communication refresher training programs are essential for the mitigation of the aforementioned barriers, and are expected to lead to the development of new strategies for the overcoming of these communication barriers.

Effective sentence-level relation extraction model using entity-centric dependency tree.

The syntactic information of a dependency tree is an essential feature in relation extraction studies. Traditional dependency-based relation extraction methods can be categorized into hard pruning methods, which aim to remove unnecessary information, and soft pruning methods, which aim to utilize all lexical information. However, hard pruning has the potential to overlook important lexical information, while soft pruning can weaken the syntactic information between entities. As a result, recent studies in relation extraction have been shifting from dependency-based methods to pre-trained language model (LM) based methods. Nonetheless, LM-based methods increasingly demand larger language models and additional data. This trend leads to higher resource consumption, longer training times, and increased computational costs, yet often results in only marginal performance improvements. To address this problem, we propose a relation extraction model based on an entity-centric dependency tree: a dependency tree that is reconstructed by considering entities as root nodes. Using the entity-centric dependency tree, the proposed method can capture the syntactic information of an input sentence without losing lexical information. Additionally, we propose a novel model that utilizes entity-centric dependency trees in conjunction with language models, enabling efficient relation extraction without the need for additional data or larger models. In experiments with representative sentence-level relation extraction datasets such as TACRED, Re-TACRED, and SemEval 2010 Task 8, the proposed method achieves F1-scores of 74.9%, 91.2%, and 90.5%, respectively, which are state-of-the-art performances.

Security risk assessment of submerged floating tunnel based on fault tree and multistate fuzzy Bayesian network

To assess the global security risks of the submerged floating tunnel (SFT) in marine environments during operation and provide a basis for risk control, a security risk assessment method using a multistate fuzzy Bayesian network (MFBN) considering complex disaster-inducing factors is proposed. A fault tree model of SFT security risk is established to analyze the causal relationships between global risk and influence factors such as structural components and environmental loads. For root nodes, fuzzy probabilities for each state are obtained through expert knowledge. An improved similarity aggregation method is proposed to integrate expert opinions, mitigating the impact of significant option discrepancies. For non-root nodes, the Leaky Noisy-Max model is used to calculate complex conditional probabilities within the SFT. The probabilities of various security risk states and key risk factors could be determined through reasoning by MFBN. Additionally, a risk prediction method that incorporates domain expert opinions and leverages the BN's ability of updating node probabilities with new information was developed to forecast the security risks of the SFT under wave and current loads.

Tank pool fire domino effect prevention by inherently safer layout planning: A techno-economic analytical index

Tank pool fire domino effects, represented by their capacities of accident consequence escalation and extensive damages, present substantial safety challenges within chemical storage systems. In this work, the inherent safety concept is introduced to proactively address these challenges while ensuring cost-effectiveness. Accordingly, a dedicated novel metric called Inherent Safety & Economic Risk Index (IS&ERI) is proposed to make safety and cost trade-offs for indicating the risk levels of various tank farm layout options. To develop the IS&ERI, the Dow's loss estimate procedure is adapted for representing the economic implications, and the Bayesian Networks are employed for indicating the failure probabilities of target units considering the synergistic effects of multiple pool fires. Two improved graph theory metrics are presented to determine the root nodes of the Bayesian Networks. For demonstrating the proposed IS&ERI, a case study is conducted and the results show that the IS&ERI of 4 tank farm layout options are 0.2991, 0.3120, 0.3525, and 0.2285, implying that Option 4 is the best layout with the lowest potential loss in case of the pool fire domino effects. As the key research contribution, the IS&ERI is presented and it can be used as a useful tool to compare and select the best tank farm layout option to eliminate or significantly reduce the knock-on effects of tank pool fires at their sources.

Quantitative Definition of Low-Health-Interest Populations by Using Regression Trees: A Nationwide Internet Survey in Japan.

Reducing health disparities is a public health issue. Identification of low-health-interest populations is important, but a definition of people with low health interest has not yet been established. We aimed to quantitatively define low-health-interest populations. A nationwide cross-sectional internet survey was conducted in 2022. We compiled regression tree (RT) analyses with/without adjustment for age, sex, and socioeconomic status with the 12-item Interest in Health Scale (IHS, score range 12-48) as an explanatory variable and the 10 composite health behaviors as a dependent variable. We defined the first IHS branching condition from the root node as a lower-health-interest group and the terminal node with the lowest health behaviors as the lowest-health-interest group. The mean IHS value of 22,263 analyzed participants was 32.1 ± 5.6; it was higher in females and in those who were aged over 45 years, had a high education, a high income, or a spouse. The first branching condition was IHS 31.5, and the terminal node branched at 24.5, before/after adjustment for covariates. We determined the cutoff values of the IHS as <32 for a lower-health-interest group and <25 for the lowest-health-interest group. Using these cutoffs might enable us to reveal the characteristics of low-health-interest populations.

IMPLEMENTATION OF THE DECISION TREE METHOD PREDICTING STUDENT GRADUATION ON TIME WITH C5.0 ALGORITHM

The percentage of students who graduate on time is one of the benchmarks for assessing a university's accreditation. To get good accreditation results, it is necessary to predict the graduation rate using a method so that students can anticipate delays in graduating. The C5.0 algorithm is a Decision Tree based algorithm which is a refinement of the ID3 and C4.5 algorithms created by Ross Quinlan in 1987. In this research, researchers analyzed several variables used in the input variables, there are several variables used, namely the SKS variable, Academic Leave, Student Status, Relationship/Communication with Family, Interests and Talents, Family Economics, Relationship/Communication with Lecturers, Number of Revisions, Final Assignment Material or Method. Meanwhile, in the target variable there is the Study Period. The aim of this research is to obtain the results of the implementation of graduation predictions for Strata 1 (S1) Mathematics Study Program students, Faculty of Science and Technology, UIN North Sumatra Medan using the C5.0 Decision Tree Algorithm. There are several steps taken, namely data preprocessing, data division, calculating information gain for the entire case, calculating information gain and entropy for each attribute. Calculate the total entropy and Gain value of each attribute, create a branch from each root node category, if the result of the branch becomes an internal node, the process is carried out until the attribute categories cannot be split. After carrying out the analysis, the accuracy rate was 82.81%. With a high level of accuracy, it can be developed into a rule that can provide predictions or input in making decisions about whether students will graduate on time

Distributed adaptive event-triggered consensus control for high-order multi-agent systems

In this paper, a distributed adaptive event-triggered control approach is proposed for the consensus problem of high-order multi-agent systems (MASs) with unknown dynamics and external disturbances. First, to save communication resources, the event-triggered control is proposed to ensure the consensus of MASs, where the trigger condition is established based on the neighbors’ information. The approximation properties of neural networks are used to address the unknown dynamics and external disturbances. Second, rigorous theoretical analysis proves that the MASs can achieve the leader-following consensus, as long as the communication graph of MASs contains a spanning tree with the leader as the root node. The filtered tracking errors and adaptive parameters estimated errors are proved to be uniformly ultimately bounded without Zeno behavior. Finally, the simulations and experiments illustrate the feasibility and effectiveness of the theoretical results.

DeepTree: Modeling Trees With Situated Latents.

In this article, we propose DeepTree, a novel method for modeling trees based on learning developmental rules for branching structures instead of manually defining them. We call our deep neural model "situated latent" because its behavior is determined by the intrinsic state -encoded as a latent space of a deep neural model- and by the extrinsic (environmental) data that is "situated" as the location in the 3D space and on the tree structure. We use a neural network pipeline to train a situated latent space that allows us to locally predict branch growth only based on a single node in the branch graph of a tree model. We use this representation to progressively develop new branch nodes, thereby mimicking the growth process of trees. Starting from a root node, a tree is generated by iteratively querying the neural network on the newly added nodes resulting in the branching structure of the whole tree. Our method enables generating a wide variety of tree shapes without the need to define intricate parameters that control their growth and behavior. Furthermore, we show that the situated latents can also be used to encode the environmental response of tree models, e.g., when trees grow next to obstacles. We validate the effectiveness of our method by measuring the similarity of our tree models and by procedurally generated ones based on a number of established metrics for tree form.

BrepGen: A B-rep Generative Diffusion Model with Structured Latent Geometry

This paper presents BrepGen , a diffusion-based generative approach that directly outputs a Boundary representation (B-rep) Computer-Aided Design (CAD) model. BrepGen represents a B-rep model as a novel structured latent geometry in a hierarchical tree. With the root node representing a whole CAD solid, each element of a B-rep model (i.e., a face, an edge, or a vertex) progressively turns into a child-node from top to bottom. B-rep geometry information goes into the nodes as the global bounding box of each primitive along with a latent code describing the local geometric shape. The B-rep topology information is implicitly represented by node duplication. When two faces share an edge, the edge curve will appear twice in the tree, and a T-junction vertex with three incident edges appears six times in the tree with identical node features. Starting from the root and progressing to the leaf, BrepGen employs Transformer-based diffusion models to sequentially denoise node features while duplicated nodes are detected and merged, recovering the B-Rep topology information. Extensive experiments show that BrepGen advances the task of CAD B-rep generation, surpassing existing methods on various benchmarks. Results on our newly collected furniture dataset further showcase its exceptional capability in generating complicated geometry. While previous methods were limited to generating simple prismatic shapes, BrepGen incorporates free-form and doubly-curved surfaces for the first time. Additional applications of BrepGen include CAD autocomplete and design interpolation. The code, pretrained models, and dataset are available at https://github.com/samxuxiang/BrepGen.

A multi-agent learning framework for mixed-integer linear programming

Mixed integer linear programming (MILP) is an important problem in the combinatorial optimization domain, which has wide applications in practical optimization scenarios. Given that most MILP problems fall into the NP-hard category, which the traditional methods may fail to solve, recent research has tried to derive MILP solutions using machine learning techniques. The whole MILP-solving procedure involves lots of modules, such as pre-solving, cut selection, node section, etc., and these modules are closely related and influence each other. However, the previous machine learning-based approaches neglect the connections between these modules, and focus on single-module learning techniques. To address this, we propose an initial step towards a more comprehensive multi-agent learning framework that allows different modules to interact and collaborate. Specifically, our current implementation involves two key modules: HEM for cut selection applied at the root node and GCNN for variable selection. By employing HEM to influence the training of GCNN, these two agents thus work in unison. Through extensive experiments on four MILP datasets in diverse scenarios, we observe significant improvements in solving time and PD integral metrics compared with the state-of-the-art learning-based MILP solving methods. This work lays the groundwork for future development of a fully integrated multi-agent framework.

RPL*: An Explainable AI-based routing protocol for Internet of Mobile Things

The Internet of Mobile Things (IoMT) is an emerging paradigm of Internet of Things (IoT) with special focus on enabling mobility to the ‘things’. Several IoMT applications such as group of robots or drones performing collaborative search and rescue operation, identification of mines, warehouse management, goods delivery, etc can be considered as examples of IoMT systems. In the applications mentioned above, the nodes may send the information in a multi-hop manner to the root or coordinator node which may be static or mobile. While the Routing Protocol for Low Power and Lossy Networks (RPL) is extensively utilized in static IoT networks, it encounters significant limitations in handling mobility and providing resilience against routing attacks in mobile IoT networks. In this work, we propose a modified RPL, RPL* which is robust to handling mobility in nodes and is resilient towards routing attacks. In RPL*, any deviation from the normal behaviors of the network are identified as anomalies using an unsupervised Explainable Artificial Intelligence (XAI) strategy. In RPL*, we propose a novel mobility detection mechanism that will identify the mobility in the network in an energy efficient manner without incurring additional communication overhead. To maintain the connectivity with parent node, we propose a novel proactive connectivity management mechanism in RPL* which will ensure a smooth transition from one parent to another if required, thus avoiding the network partitioning due to mobility. The performance analysis of the system has demonstrated an improvement in packet delivery ratio of the mobile nodes by 40% due to the proposed RPL* when compared to RPL. Also, the proposed XAI strategy provided an F1-score of over 95% for the detection of sink hole and black hole attacks in the tested IoMT network scenarios. It was observed that RPL* improves the performance of the IoMT network when compared to RPL. However it may be noted that the mechanisms introduced to support mobility does not lead to a drop in PDR or increase in control packet overhead for static networks. Hence, RPL* can be considered as an alternative to RPL for IoT as well as IoMT networks.

Design of consensus and cluster consensus controller for first-order nonlinear multi-agent systems based on subgroup structure

It is well known that nonlinear multi-agent systems (MASs) with a directed topology can reach a consensus if the global coupling coefficient is strong enough. In this paper, we propose a novel distributed protocol based on the subgroup structure. Our results demonstrate that the MAS design can follow a part-to-whole approach, starting with subgroups and gradually building up to a more complex system. We firstly consider two cases for the MASs: one with a single leader and one with a leader group, where agents within the leader group can communicate with each other, but cannot receive the information of other group. We show that followers can track the leader or group leaders if intra-coupling within the subgroup and information strength received by the root node are above a certain threshold. Next, we investigate cluster consensus and practical cluster consensus for the Laplacian matrix with unequal row sum, equal row sum, and mixing of partitioned matrices. Our results reveal that row sum and community structure of the partitioned matrix are crucial in determining the final state of each agent. Our proposed control protocol is highly flexible and relies only on the cluster structure and information received by the root node, rather than global information. Moreover, our method is not limited to tracking problems, cooperative control, or multi-objective control, and can accommodate new group agents joining the system, making it highly adaptable. We provide illustrative examples to demonstrate the effectiveness of our approach.

Semi-supervised clustering guided by pairwise constraints and local density structures

Clustering based on local density peaks and graph cut (LDP-SC) is one of the state-of-the-art algorithms in unsupervised clustering, which first divides the data set to be multiple local trees, and then aggregates these local trees to obtain the final clustering result. However, for complex data sets, there might exist data points from different classes in the same local tree. In this article, we use pairwise constraint information to resolve this issue and propose a semi-supervised local density peaks and graph cut based clustering algorithm (SLDPC). In particular, SLDPC proposes intra-cluster conflict resolution and inter-cluster conflict resolution steps to split the local trees which are inconsistent with the provided pairwise constraint information. Theoretically, we show that the two steps will finish in a finite number of operations and the split local trees will be consistent with the pairwise constraint information. Subsequently, root node redirection and noise filtering steps are designed to avoid the local trees becoming too fragmented. Finally, we exploit the E2CP algorithm to further improve the similarity matrix between local trees using the pairwise constraint information, and the spectral clustering algorithm is adopted to obtain the clustering result. Experiments on multiple widely used synthetic and real-world data sets show that SLDPC is superior to LDP-SC and several other semi-supervised prominent clustering algorithms for most of the cases.

Integrated management of safety and security barriers in chemical plants to cope with emerging cyber-physical attack risks under uncertainties

Chemical facilities face threats from accidental and intentional events, including the rising concern of cyber-physical (C2P) attacks in the digitized industrial control system era. Addressing major accident risks from safety hazards and C2P attacks requires an immediate unified framework for safety and security barrier management. This study presents a systematic risk-based approach to integrate conventional safety risks with emerging C2P attack risks. Adverse scenarios are identified, integrated into an attack-tree-bow-tie diagram, and modelled using a Bayesian network (BN). A vulnerability assessment model is developed to quantify industrial control system vulnerability to C2P attacks, considering uncertainties in attackers' knowledge levels. Monte Carlo simulations are used to handle uncertainty propagation in risk assessment, allowing the use of probability distributions for BN root nodes. Sensitivity analysis identifies critical factors/events, guiding the proposal of candidate strategies for barrier improvements. Combining cost-effectiveness analysis with a risk matrix yields the optimal strategy for safety and security barrier enhancements based on risk estimations. A hypothetical case study demonstrates the proposed approach's effectiveness in integrated safety and security barrier management, considering security vulnerability patching and safety barrier maintenance scheduling from a cost-effective perspective.

Response Phaseolus vulgaris L. Plant to Glomus mosseae and Azotobacter chroococcum

An experiment was performed to study the impact of Mycorrhizal fungus (Glomus mosseae) and Azotobacter chroococcum on the growth parameters of the common bean (Phaseolus vulgaris L.) plant. The experimental methodology employed in this study was a factorial design, specifically a randomized complete block design (RCBD), with three replications. The treatments consisted of four different biological fertilizers, namely: none-inoculated, inoculated with arbuscular mycorrhizal fungi (AMF), inoculated with Azotobacter, and a mixture of both (AMF+Azotobacter). The use of biofertilizers, whether used singly or in combination, resulted in improved germination rates, plant height, number of root nodes, as well as fresh and dry weights of both shoots and roots particularly (AMF + Azoto) treatment (96.33±1.52, 37.33±2.08, 50.00±2.00, 8.83±0.40, 3.90±0.20, 6.53±0.23, 2.30±0.20) respectively, compared to plants that were not infected. In addition, the dual treatment of the fungus AMF and the bacteria A. chroococcum was recorded a maximum mycorrhizal infection percentage of 90% compared with the control which recorded 20%. The results of this study suggest the existence of an additive or a synergistic relationship between Glomus mosseae and A.chroococcum.

Analyzing impact of network constraints on feasible operation region of the radial distribution networks

Recently, there has been lot of research on finding the feasible operating region (FOR) of the active distribution network (ADN) to provide ancillary services to the transmission grid. The current focus of research is on finding methods that can accurately determine the shape of the FOR and have fast computation time. However, there is little focus on analyzing how network constraints impact the shape of the FOR. This paper attempts to address this issue and provides a mathematical derivation of how network constraints, such as cable loading and voltage limits, would influence FOR. The approach is based on a linear distflow mathematical analysis. The relations derived in the analysis are verified with numerical simulation on radial low-voltage (LV) and medium-voltage (MV) networks. The results show that cable loading constraints cut the FOR in a circular way and voltage constraints cut the FOR with a straight line of slope −ϵ, which is the ratio Rn/Xn of the path from root node to the node under analysis of the radial distribution network.

A Branch-and-Price algorithm for the electric autonomous Dial-A-Ride Problem

The Electric Autonomous Dial-A-Ride Problem (E-ADARP) consists in scheduling a fleet of electric autonomous vehicles to provide ride-sharing services for customers that specify their origins and destinations. The E-ADARP considers the following perspectives: (i) a weighted-sum objective that minimizes both total travel time and total excess user ride time; (ii) the employment of electric autonomous vehicles and a partial recharging policy. This paper presents the first labeling algorithm for a path-based formulation of the DARP/E-ADARP, where the main ingredient includes: (1) fragment-based representation of paths, (2) a novel approach that abstracts fragments to arcs while ensuring excess-user-ride-time optimality, (3) construction of a sparser new graph with the abstracted arcs, which is proven to preserve all feasible routes of the original graph, and (4) strong dominance rules and constant-time feasibility checks to compute the shortest paths efficiently. This labeling algorithm is then integrated into Branch-and-Price (B&P) algorithms to solve the E-ADARP. In the computational experiments, the B&P algorithm achieves optimality in 71 out of 84 instances. Remarkably, among these instances, 50 were solved optimally at the root node without branching. We identify 26 new best solutions, improve 30 previously reported lower bounds, and provide 17 new lower bounds for large-scale instances with up to 8 vehicles and 96 requests. In total 42 new best solutions are generated on previously solved and unsolved instances. In addition, we analyze the impact of incorporating the total excess user ride time within the objectives and allowing unlimited visits to recharging stations. The following managerial insights are provided: (1) solving a weighted-sum objective function can significantly enhance the service quality, while still maintaining operational costs at nearly optimal levels, (2) the relaxation on charging visits allows us to solve all instances feasibly and further reduces the average solution cost.

Multi-scale sampling attention graph convolutional networks for skeleton-based action recognition

Skeleton-based action recognition has attracted increasing interest in recent years. With the flexibility of modeling long-range dependency of joints, the self-attention module has served as the basic component in skeleton-based action recognition. However, the global receptive field of self-attention is not conducive to the modeling of skeleton locality, and the self-attention model is imbued with less inductive bias, which leads to overfitting. In this paper, we propose an attention graph convolutional network (AGCN) with multi-scale sampling to effectively model the local and global features of the skeleton. Firstly, we propose two extreme sampling strategies for generating and ordering neighboring nodes of root nodes. A local-first sampling method is introduced to construct local graph windows, and a global-first sampling method is proposed to assemble long-range joints for constructing global graph windows. The local-first sampling and global-first sampling introduce more skeleton-specific inductive biases to regularize the model capacity. Secondly, the AGCN combines the self-attention mechanism with graph convolution operation, which alleviates the over-smoothing of graph convolution and preserves the translation invariant. Based on the multi-scale sampling strategy, the AGCN can effectively model the locality and non-locality of the skeleton. Finally, by coupling the aforementioned proposals, we develop a two-pathway model for multi-scale feature fusion. Extensive experiments demonstrate that our model could achieve comparable performance with state-of-the-art works on the NTU RGB+D 60, NTU RGB+D 120, the UAV-HUMAN and NW-UCLA datasets.