Depth-first Research Articles

Ant Colony Optimization for Accelerated Pathway Identification in Connection Element Method Reservoir Models: A Fast-Track Solution for Large-Scale Simulations

In recent years, reservoir models based on the Connection Element Method (CEM) have gained extensive application in reservoir development. This mesh-free modeling approach effectively captures all flow paths and flow-splitting coefficients between nodes, providing a clear view of flow interactions and accurately identifying primary connectivity pathways between injection and production wells. However, the traditional approach of traversing flow paths and splitting coefficients imposes a significant computational load, particularly when applied to large reservoirs with numerous virtual wells. To enhance simulation efficiency, this paper introduces a novel method leveraging the Ant Colony Optimization (ACO) algorithm to efficiently identify the path with the highest splitting coefficient between well pairs. This approach rapidly calculates and filters the dominant connectivity paths between injection and production wells in CEM models. A comparative analysis shows that, while the ACO algorithm provides limited benefit with a small number of connectivity paths, it significantly outperforms the conventional depth-first search algorithm as the number of experimental wells increases.

Optimization of Cast-Steel Tubular Circular-Hollow-Section Connections Based on Depth-First Search Algorithm

This study introduces a novel design for a cast-steel joint in the shape of a T, aimed at resolving concerns regarding stress concentration at points where geometric intersections occur and enhancing the quality of welding in T-shaped welded joints. The proposed integrated design framework greatly facilitated the successful construction of a three-dimensional joint between a brace and a chord at a T-shaped node. The geometric parameters of curves in the connector were optimized using a depth-first search algorithm, resulting in control points for the optimized curve. Computer-aided design software was then employed to obtain the refined connector. The design framework has the ability to produce designs with smooth and uninterrupted boundaries, making them highly compatible with traditional casting methods and effectively tackling the manufacturing challenges related to topology optimization. The numerical simulation results demonstrate that, in comparison to traditionally welded T-joints of the same size, the stress concentration factor of the optimized joints exhibits a significant reduction, accompanied by a notable disparity in stress distribution. Moreover, the impact of the thickness of the brace and the axial compression ratio on the stress concentration factor of the optimized joints was relatively insignificant. The stress concentration factor of the cast-steel joint was reduced by more than 84%, leading to a significant enhancement in fatigue performance.

Face-Gaze Biometric Fusion for Enhanced Personal Authentication

Objectives: To develop a robust authentication system by combining two authentication modals. The system aims to enhance the security by integrating the gaze-based biometrics in the authentication process, which is difficult to perform spoofing attacks. Methods: The system uses gaze-tracking and face recognition modals to authenticate the user. The user will enter his/her credentials, then the first step of authentication will start where the face recognition modal will authenticate the user by detecting the user’s face, if one is not authenticated in this phase then the user will be omitted. After that the second phase will starts where the system will generate a random 10X10 maze using Depth-First Search algorithm, and then it will extract the x, y coordinates of the maze. The system will capture the user’s real-time eye movement data, and extracts the x, y coordinates of the pupil through a webcam, and compares those coordinates with the maze coordinates if both are matched then the system will provide access to the user else it will omit the user. Findings: Findings show that this approach improves the security of traditional biometric systems, offering a unique form of user authentication. The system tested for accuracy in different environments, and its performance was evaluated based on how well the user’s gaze matches the maze path. Tests across various environments show that the system achieves an accuracy of 92% in matching user gaze to the maze coordinates, also achieved 0% FTE (Failure To Enroll) and 0% FTA (Failure To Acquire). Novelty: The novelty of this project lies in combining biometric eye gaze tracking with real-time maze-based navigation for user authentication. Unlike the traditional biometrics that rely on static features like fingerprints or facial recognition, this system incorporates dynamic tracking of user behavior, specifically gaze patterns, to enhance security. This makes it more resilient to spoofing attacks. Keywords: Biometric Authentication, Deepfake Resistance, Authorization Mechanism, Operational Environments, Human-Centric Security

Adaptive Emergency Evacuation Routing: A Graph-Based Approach

Emergency evacuation plays a vital role in disaster management operations. The existing solutions for planning and routing emergency evacuations rely on preplanning based on prior information and lack adaptability to unprecedented conditions. This study introduces a novel adaptive method for emergency evacuation routing that dynamically reshapes the road network by actively determining and adjusting the direction of road segments and assigning traffic flow to them in order to improve the maximum possible flow of traffic evacuated from a set of dangerous zones to a set of safe zones. The proposed adaptive method revolves around a novel dynamic graph-based algorithm that iteratively and recursively distributes the traffic flow across the road network using a Depth First Search (DFS) approach. When the algorithm assigns the traffic flow to the road segments, it also determines the direction of the road segments to dynamically improve the performance of the evacuation process. The proposed method was implemented in a real-world evacuation scenario in the road network of Hoboken, New Jersey. The outcomes indicated a significant improvement of more than 74% in the number of successfully evacuated vehicles when the proposed algorithm was used, compared to state-of-the-art methods for evacuation routing that are unable to adjust road directions. The outcomes of this study help decision-makers and first responders develop dynamic and adaptive emergency evacuation plans for successful disaster management operations.

Depth-First Search and Research of Algorithms Related to Bipartite Graphs

The application of bipartite graph is extensively used in many fields of daily production and life. This in turn led to a lot of research and thinking about the bipartite graph: what is a bipartite graph? How to tell if a graph is a bipartite graph? And related algorithms for some specific characteristics of a bipartite graph. This paper uses the literature research method and the experiment method, starting from the depth-first-search, particularly summarizes and clearly explains the related algorithms of the bipartite graph: the judgment of the bipartite graph, and the calculation of its maximum matching number. Lays a foundation for solving more problems that require the application of bipartite graphs.

Comparative analysis of AI-based search algorithms in solving 8 puzzle problems

BackgroundThe 8-puzzle problem is a well-known combinatorial search problem, often used to test the effectiveness of various artificial intelligence (AI) algorithms. This study aims to evaluate the performance of three AI-based search algorithms—Breadth-First Search (BFS), Depth-First Search (DFS), and A* Search—in solving the 8-puzzle problem. The algorithms were implemented in Java, and their performance was measured in terms of solution length, the number of expanded nodes, and execution time.ResultsExperiments were conducted on eight randomly generated instances of the 8-puzzle problem. The findings reveal that both BFS and A* Search consistently identify optimal solutions in all cases. However, BFS was found to use more memory and operate slower than A* Search. Conversely, DFS demonstrated faster execution times and lower memory usage compared to BFS but was less reliable in finding optimal or feasible solutions. The study also explored the impact of different heuristic functions on the performance of A* Search. Among the heuristics tested, the Manhattan distance heuristic was determined to be the most effective, offering the best balance between accuracy and efficiency.ConclusionsIn conclusion, both BFS and A* Search are effective in finding optimal solutions, with A* Search being more efficient in terms of speed and memory usage. DFS, while faster and more memory-efficient, is less consistent in producing optimal solutions. The Manhattan distance heuristic significantly enhances the performance of A* Search, making it the preferred heuristic for the 8-puzzle problem. These results suggest that A* Search, particularly with the Manhattan distance heuristic, is a highly efficient choice for solving the 8-puzzle problem, especially in contexts where computational resources are limited.

A Probabilistic Approach in the Search Space of the Molecular Distance Geometry Problem.

The discovery of the three-dimensional shape of protein molecules using interatomic distance information from nuclear magnetic resonance (NMR) can be modeled as a discretizable molecular distance geometry problem (DMDGP). Due to its combinatorial characteristics, the problem is conventionally solved in the literature as a depth-first search in a binary tree. In this work, we introduce a new search strategy, which we call frequency-based search (FBS), that for the first time utilizes geometric information contained in the protein data bank (PDB). We encode the geometric configurations of 14,382 molecules derived from NMR experiments present in the PDB into binary strings. The obtained results show that the sample space of the binary strings extracted from the PDB does not follow a uniform distribution. Furthermore, we compare the runtime of the symmetry-based build-Up (SBBU) algorithm (the most efficient method in the literature to solve the DMDGP) combined with FBS and the depth-first search (DFS) in finding a solution, ascertaining that FBS performs better in about 70% of the cases.

Research on Path Planning of Agricultural UAV Based on Improved Deep Reinforcement Learning

Traditional manual or semi-mechanized pesticide spraying methods often suffer from issues such as redundant coverage and cumbersome operational steps, which fail to meet current pest and disease control requirements. Therefore, there is an urgent need to develop an efficient pest control technology system. This paper builds upon the Deep Q-Network algorithm by integrating the Bi-directional Long Short-Term Memory structure to propose the BL-DQN algorithm. Based on this, a path planning framework for pest and disease control using agricultural drones is designed. This framework comprises four modules: remote sensing image acquisition via the Google Earth platform, task area segmentation using a deep learning U-Net model, rasterized environmental map creation, and coverage path planning. The goal is to enhance the efficiency and safety of pesticide application by drones in complex agricultural environments. Through simulation experiments, the BL-DQN algorithm achieved a 41.68% improvement in coverage compared with the traditional DQN algorithm. The repeat coverage rate for BL-DQN was 5.56%, which is lower than the 9.78% achieved by the DQN algorithm and the 31.29% of the Depth-First Search (DFS) algorithm. Additionally, the number of steps required by BL-DQN was only 80.1% of that of the DFS algorithm. In terms of target point guidance, the BL-DQN algorithm also outperformed both DQN and DFS, demonstrating superior performance.

Integrating hybrid deep learning and path allocation for real-time inbound passenger flow prediction and anomaly detection in urban rail transit

This paper study the problem of real-time prediction of inbound passenger flow and the detection and alerting of abnormal passenger flows in urban rail transit (URT) networks. We propose a fused framework that combines a hybrid deep learning model and an evaluation strategy. Specifically, the learning model incorporates Graph Convolutional Networks (GCN), Gated Recurrent Units (GRU), and attention mechanisms to effectively capture spatial and temporal correlations in passenger flow data. The evaluation strategy utilizes a depth-first search algorithm to determine the optimal travel paths for each individual passenger. And based on the paths, we develop a real-time method for estimating the origin–destination (OD) matrix that utilizes both long-term and short-term historical destination trend vectors to reduce dimensions while improving predictive accuracy. Through extensive testing using data from the Shanghai rail transit system, we demonstrate that this fused framework achieves high prediction accuracy for inbound passenger flow at various stations while efficiently identifying and warning sudden large-scale events involving significant increases in passenger flow volume. This research contributes towards improving overall passenger experience as well as operational resilience within urban rail systems when dealing with large-scale influxes of passengers.

Machine learning-assisted design of flow fields for proton exchange membrane fuel cells

Optimizing the flow field is a key approach to enhancing the performance of proton exchange membrane fuel cells. Most previous research on flow field design relies on physical intuitions, lacking systematic exploration of the complex geometries of flow fields. To address these issues, we first generate a comprehensive flow field library containing 28,348 geometries using the Depth-First Search algorithm. Subsequently, we randomly select 480 flow fields from this library and characterize their corresponding fuel cell performance using computational fluid dynamics. These 480 flow fields serve as the dataset for machine learning training. Using the trained neural network, we rapidly predict fuel cell performance and identify high-performance flow fields. Simulation results demonstrate that the predicted high-performance flow fields effectively improve mass transfer and current density distribution, thereby enhancing current density and maximum power density. Experimental validation shows a 10.37 % increase in maximum power density for our optimized flow field design compared to traditional serpentine channels. Additionally, our geometric analysis identifies key features of high-performance flow fields, guiding future designs.

Improving community detection in social networks using enhanced BSO by exploring network structure

Community detection in social networks plays a crucial role in understanding various social phenomena. However, accurately identifying communities is challenging due to the complex nature of social networks. To address this challenge, we propose CD-BSO, an innovative metaheuristic based on brainstorming. CD-BSO leverages network knowledge through specialized initialization and solution generation operators designed to capture social network characteristics. In the initialization phase, a technique combining Depth-First Search (DFS) is employed to consider both connectivity and information diffusion, ensuring accurate community representation. The generation step introduces search operators that consider link formation and node similarity, facilitating convergence towards a community structure that closely resembles the network's actual structure. The evaluation of CD-BSO outperforms recent and well-known community detection methods. The evaluation results were further analyzed using visual analysis, providing valuable insights into the network structure. CD-BSO exhibits significant potential for accurately identifying communities and extracting meaningful information from social networks.

GEMimp: An Accurate and Robust Imputation Method for Microbiome Data Using Graph Embedding Neural Network

Microbiome research has increasingly underscored the profound link between microbial compositions and human health, with numerous studies establishing a strong correlation between microbiome characteristics and various diseases. However, the analysis of microbiome data is frequently compromised by inherent sparsity issues, characterized by a substantial presence of observed zeros. These zeros not only skew the abundance distribution of microbial species but also undermine the reliability of scientific conclusions drawn from such data. Addressing this challenge, we introduce GEMimp, an innovative imputation method designed to infuse robustness into microbiome data analysis. GEMimp leverages the node2vec algorithm, which incorporates both Breadth-First Search (BFS) and Depth-First Search (DFS) strategies in its random walks sampling process. This approach enables GEMimp to learn nuanced, low-dimensional representations of each taxonomic unit, facilitating the reconstruction of their similarity networks with unprecedented accuracy.Our comparative analysis pits GEMimp against state-of-the-art imputation methods including SAVER, MAGIC and mbImpute. The results unequivocally demonstrate that GEMimp outperforms its counterparts by achieving the highest Pearson correlation coefficient when compared to the original raw dataset. Furthermore, GEMimp shows notable proficiency in identifying significant taxa, enhancing the detection of disease-related taxa and effectively mitigating the impact of sparsity on both simulated and real-world datasets, such as those pertaining to Type 2 Diabetes (T2D) and Colorectal Cancer (CRC). These findings collectively highlight the strong effectiveness of GEMimp, allowing for better analysis on microbial data. With alleviation of sparsity issues, it could be greatly facilitated in downstream analyses and even in the field of microbiology.

Memoization in Model Checking for Safety Properties with Multi-Swarm Particle Swarm Optimization

In software engineering, errors or faults in software systems often lead to critical social problems. One effective methodology to tackle this problem is model checking, which is an automated formal verification technique. In traditional model checking, the task of finding specification errors is reduced to deterministic search techniques such as Depth-First Search. Recent research has shown that swarm intelligence offers a powerful search capability compared to traditional techniques. In particular, multi-swarm Particle Swarm Optimization is known to be efficient and can mitigate the state-space explosion problem, i.e., the exponential increase in the search space with a linear increase in the problem size. However, the state-space explosion problem is still significant when verifying very large systems. Further performance improvement is needed. To achieve this, we propose a novel memoization or cache mechanism for storing tentative solutions for reuse in the later stages of the search procedure. For each stage, a candidate solution computed by a swarm is summarized efficiently and heuristically to consolidate similar solutions into a single representative solution. We store the summary and its associated solutions in key-value maps. Instead of computing known solutions repeatedly, we retrieve the solution if the stored key matches the summary. We incorporated the proposed mechanism into a model-checking technique with multi-swarm Particle Swarm Optimization and evaluated the search performance. We show in this paper that the proposed mechanism improved time and space consumption while maintaining solution quality.

NUMA-aware parallel sparse LU factorization for SPICE-based circuit simulators on ARM multi-core processors

In circuit simulators that resemble the Simulation Program with Integrated Circuit Emphasis (SPICE), one of the most crucial steps is the solution of numerous sparse linear equations generated by frequency domain analysis or time domain analysis. The sparse direct solvers based on lower-upper (LU) factorization are extremely time-consuming, so their performance has become a significant bottleneck. Despite the existence of some parallel sparse direct solvers for circuit simulation problems, they remain challenging to adapt in terms of performance and scalability in the face of rapidly evolving parallel computers with multiple NUMA hardware based on ARM architecture. In this paper, we introduce a parallel sparse direct solver named HLU, which re-examines the performance of the parallel algorithm from the viewpoint of parallelism in pipeline mode and the computing efficiency of each task. To maximize task-level parallelism and further minimize the thread waiting time, HLU devises a fine-grained scheduling method based on an elimination tree in pipeline mode, which employs depth-first search (DFS-like) to iteratively search for parent tasks and then place dependent tasks in the same task queue. HLU also suggests two NUMA node affinity strategies: thread affinity optimization based on NUMA nodes topology to guarantee computational load balancing and data affinity optimization to enable effective memory placement when threads access data. The rationality and effectiveness of the sparse solver HLU are validated by the SuiteSparse Matrix Collection. In comparison with KLU and NICSLU, the experimental results and analysis show that HLU attains a speedup of up to 9.14× and 1.26x (geometric mean) on a Huawei Kunpeng 920 Server, respectively.

CLIP-DFGS: A Hard Sample Mining Method for CLIP in Generalizable Person Re-Identification

Recent advancements in pre-trained vision-language models like CLIP have shown promise in person re-identification (ReID) applications. However, their performance in generalizable person re-identification tasks remains suboptimal. The large-scale and diverse image-text pairs used in CLIP's pre-training may lead to a lack or insufficiency of certain fine-grained features. In light of these challenges, we propose a hard sample mining method called DFGS (Depth-First Graph Sampler), based on depth-first search, designed to offer sufficiently challenging samples to enhance CLIP's ability to extract fine-grained features. DFGS can be applied to both the image encoder and the text encoder in CLIP. By leveraging the powerful cross-modal learning capabilities of CLIP, we aim to apply our DFGS method to extract challenging samples and form mini-batches with high discriminative difficulty, providing the image model with more efficient and challenging samples that are difficult to distinguish, thereby enhancing the model's ability to differentiate between individuals. Our results demonstrate significant improvements over other methods, confirming the effectiveness of DFGS in providing challenging samples that enhance CLIP's performance in generalizable person re-identification.

Safety evaluation for the dismantling of long-span spatial lattice structures based on deep learning and graph traversal

The demolition issue of long-span spatial lattice structures has become increasingly prominent with the rapid urban development and building renewal. Compared with traditional demolition techniques, building deconstruction offers superior economic and social benefits. Dismantling in blocks is one of the primary ways to realize the deconstruction of long-span spatial lattice structures, yet the safety is rarely studied. To fill this gap, a safety evaluation framework was established based on deep learning and graph traversal algorithm in this study. Hazardous configuration of the remaining structure is the basis of dismantling safety evaluation. To collect the hazardous configurations, an identification method in conjunction with deep neural network (DNN) and depth-first search (DFS) algorithm was proposed. The dismantling safety is measured by the improved structural well-formedness whose allowable range is determined by statistical analysis of hazardous configurations. The applicability of the framework was illustrated by the safety evaluation for the dismantling of single-layer reticulated domes. Results indicated that the dismantling safety is significantly affected by the configuration quality of the remaining structure, and the DNN-based prediction model is effective to classify the configuration state. Additionally, a large number of hazardous configurations were obtained through the identification method, and the allowable value of the safety evaluation index was determined to be 0.7. The configuration quality mainly depends on the geometry and boundary conditions. This novel framework provides scientific and effective methodological guidance for engineers, promoting the further application of building deconstruction.

Distinguishing between topological isomorphism and topological equivalence of power electronic converters

In the process of deducing the topology of power electronic converters, scholars often use topological equivalence or topological isomorphism to identify topologies with different structures but the same performance to avoid repeated research. However, the connotations of topological equivalence and topological isomorphism are not the same. This paper will clarify the method for accurately identifying topologies with the same performance by distinguishing the differences and connections between the two. First, it is derived that the necessary condition for topological isomorphism is that the determinants of their adjacency matrices are equal; then, it is derived that the necessary and sufficient conditions for topological equivalence are that their component composition is the same and their simple circuits correspond one to one; finally, the above two conditions are analyzed from the perspective of topological subgraphs, and it is found that topological isomorphism is a sufficient but not necessary condition for topological equivalence, and topological equivalence is a necessary and sufficient condition for the same topological performance. Therefore, in practice, equivalence rather than isomorphism should be used to identify topologies with the same performance. This paper verifies the correctness of the theoretical analysis through a case analysis. In addition, this paper also introduces a method for automatically determining equivalent topologies based on the depth-first search algorithm to help quickly and accurately identify converter topologies with the same performance.

Avoid backtracking and burn your inputs: CONUS-scale watershed delineation using OpenMP

The Memory-Efficient Watershed Delineation (MESHED) parallel algorithm is introduced for Contiguous United States (CONUS)-scale hydrologic modeling. Delineating tens of thousands of watersheds for a continental-scale study can not only be computationally intensive, but also be memory-consuming. Existing algorithms require separate input and output data stores. However, as the number of watersheds to delineate and the resolution of input data grow significantly, the amount of memory required for an algorithm also quickly increases. MESHED uses one data store for both input and output by destructing input data as processed and a node-skipping depth-first search to further reduce required memory. For 1000 watersheds in Texas, MESHED performed 95% faster than the Central Processing Unit (CPU) benchmark algorithm using 33% less memory. In a scaling experiment, it delineated 100,000 watersheds across the CONUS in 13.64s. Given the same amount of memory, MESHED can solve 50% larger problems than the CPU benchmark algorithm can.

Analysis, Generation, and Validation of New Boards for the Game Micro Robots

This document presents a study of the board game Micro Robots and the development of a pipeline for the generation and validation of new boards of variable sizes. The implementation was carried out in the C# language, using graph theory and a depth-first search algorithm to explore various board configurations and their possible solutions. The main objective is the creation of boards that are different from the original ones, offering greater variety in the gaming experience. The impact of the modifications on the game’s dynamics and complexity was evaluated by comparing the generated boards with those from the commercial game. The results show that no major differences were found between the new boards’ structure and the original ones, maintaining an average complexity across all configurations. This work not only contributes to the study of board game design, but also opens new opportunities for innovation in this field.

Making Formulog Fast: An Argument for Unconventional Datalog Evaluation

With its combination of Datalog, SMT solving, and functional programming, the language Formulog provides an appealing mix of features for implementing SMT-based static analyses (e.g., refinement type checking, symbolic execution) in a natural, declarative way. At the same time, the performance of its custom Datalog solver can be an impediment to using Formulog beyond prototyping—a common problem for Datalog variants that aspire to solve large problem instances. In this work we speed up Formulog evaluation, with some surprising results: while 2.2× speedups can be obtained by using the conventional techniques for high-performance Datalog (e.g., compilation, specialized data structures), the big wins come by abandoning the central assumption in modern performant Datalog engines, semi-naive Datalog evaluation. In the place of semi-naive evaluation, we develop eager evaluation, a concurrent Datalog evaluation algorithm that explores the logical inference space via a depth-first traversal order. In practice, eager evaluation leads to an advantageous distribution of Formulog’s SMT workload to external SMT solvers and improved SMT solving times: our eager evaluation extensions to the Formulog interpreter and Soufflé’s code generator achieve mean 5.2× and 7.6× speedups, respectively, over the optimized code generated by off-the-shelf Soufflé on SMT-heavy Formulog benchmarks. All in all, using compilation and eager evaluation (as appropriate), Formulog implementations of refinement type checking, bottom-up pointer analysis, and symbolic execution achieve speedups on 20 out of 23 benchmarks over previously published, hand-tuned analyses written in F ♯ , Java, and C++, providing strong evidence that Formulog can be the basis of a realistic platform for SMT-based static analysis. Moreover, our experience adds nuance to the conventional wisdom that traditional semi-naive evaluation is the one-size-fits-all best Datalog evaluation algorithm for static analysis workloads.