Data Generation Approach Research Articles

From shallows to depths: unveiling hybrid synthetic data modeling for enhanced learning with privacy considerations in naturally imbalanced datasets

Data serves as the foundational element that drives model development and performance in machine learning and deep learning. The learning algorithms are as good as data on which they are trained on. Data constrained environments pose serious threat to the effectiveness of learning algorithms. Data limitations stem from a range of factors, including regulatory restrictions, privacy concerns, and the inherent scarcity of relevant data. This constrained availability of data often leads to class imbalance problem in the context of classification tasks. To address the challenges of limited data, the algorithmic generated data, also known as synthetic data, is gaining significant traction as a cost-effective, readily available, and secure alternative. Synthetic data generation techniques can be employed to enhance dataset size by augmenting data samples and to address class imbalance by increasing the number of minority class instances. These techniques generally fall into two main categories: distance-based shallow models and probability estimation-based deep generative models. Shallow interpolation-based models generate new data points within the local space between existing data points, while deep density estimation based models generate new data by learning the whole distribution of data. . In the context of smaller datasets, deep generative models often struggle to accurately estimate the probability distribution of whole data. To effectively represent the global data distribution, these deep models require initial starting data samples to guide their approximation. This paper examines the potential of integration of shallow and deep generative models in the data generation pipeline for effective synthetic data augmentation which furthers enhanced learning and generalization of downstream tasks. In this work, we present a hybrid approach of tabular data generation involving mixed type data attributes (continuous, discrete) and pay special attention to data imbalance and insufficient data problems. We introduce the Hybrid Data Balancing and Augmentation Approach for Mixed Tabular Data (HDBA-MTD), specifically designed to synthesize samples for underrepresented labels of output class and address issues of insufficient data instances. This approach enhances training data diversity, thereby paying special attention to the downstream classification and generalization performance. Experiments are carried out using benchmark datasets to assess the practicality of the presented hybrid model in real-world scenarios. This work has also attempted to quantify the privacy preservability for real data concerning ethical considerations and data security circumstances. The evaluation and analysis of these experiments show that the present hybrid model performs favorably compared to other current hybrid synthetic data generation methods.

Generating 3D Models for UAV-Based Detection of Riparian PET Plastic Bottle Waste: Integrating Local Social Media and InstantMesh

In recent years, waste pollution has become a severe threat to riparian environments worldwide. Along with the advancement of deep learning (DL) algorithms (i.e., object detection models), related techniques have become useful for practical applications. This work attempts to develop a data generation approach to generate datasets for small target recognition, especially for recognition in remote sensing images. A relevant point is that similarity between data used for model training and data used for testing is crucially important for object detection model performance. Therefore, obtaining training data with high similarity to the monitored objects is a key objective of this study. Currently, Artificial Intelligence Generated Content (AIGC), such as single target objects generated by Luma AI, is a promising data source for DL-based object detection models. However, most of the training data supporting the generated results are not from Japan. Consequently, the generated data are less similar to monitored objects in Japan, having, for example, different label colors, shapes, and designs. For this study, the authors developed a data generation approach by combining social media (Clean-Up Okayama) and single-image-based 3D model generation algorithms (e.g., InstantMesh) to provide a reliable reference for future generations of localized data. The trained YOLOv8 model in this research, obtained from the S2PS (Similar to Practical Situation) AIGC dataset, produced encouraging results (high F1 scores, approximately 0.9) in scenario-controlled UAV-based riparian PET bottle waste identification tasks. The results of this study show the potential of AIGC to supplement or replace real-world data collection and reduce the on-site work load.

Mitigating adversarial cascades in large graph environments

A significant amount of society’s infrastructure can be modeled using graph structures, from electric and communication grids, to traffic networks, to social networks. Each of these domains are also susceptible to unwanted cascades, whether this be overloaded devices in the power grid or the reach of a social media post containing misinformation. The potential harm of a cascade is compounded when considering a malicious attack by an adversary that is intended to maximize the cascade impact. However, by exploiting knowledge of the cascading dynamics, targets with the largest cascading impact can be preemptively prioritized for defense, and the damage an adversary can inflict can be mitigated. While game theory provides tools for finding an optimal preemptive defense strategy, existing methods struggle to scale to the context of large graph environments because of the combinatorial explosion of possible actions with the size of the graph. Data-driven deep learning approaches can fill this gap, but they demand a large amount of data that is often costly to obtain. We propose a novel data generation approach for the cascading failure security problem that uses counterfactual data augmentation and a sub-game data querying strategy to generate both high quality and a high quantity of training data. We demonstrate that our data generation approach creates up to an order of magnitude more data than naïve querying with double the efficiency. We also introduce a deep learning model based embedding node pairs, and train it on our generated data, showing that it effectively defends from cascades on graphs as large as 1000 nodes. Our deep learning approach achieves a closer approximation to the Nash Equilibrium than existing heuristic methods, for graphs where the ground-truth equilibrium can be calculated.

Improving Machine Learned Force Fields for Complex Fluids through Enhanced Sampling: A Liquid Crystal Case Study.

Machine learned force fields offer the potential for faster execution times while retaining the accuracy of traditional DFT calculations, making them promising candidates for molecular simulations in cases where reliable classical force fields are not available. Some of the challenges associated with machine learned force fields include simulation stability over extended periods of time and ensuring that the statistical and dynamical properties of the underlying simulated systems are correctly captured. In this work, we propose a systematic training pipeline for such force fields that leads to improved model quality, compared to that achieved by traditional data generation and training approaches. That pipeline relies on the use of enhanced sampling techniques, and it is demonstrated here in the context of a liquid crystal, which exemplifies many of the challenges that are encountered in fluids and materials with complex free energy landscapes. Our results indicate that, whereas the majority of traditional machine learned force field training approaches lead to molecular dynamics simulations that are only stable over hundred-picosecond trajectories, our approach allows for stable simulations over tens of nanoseconds for organic molecular systems comprising thousands of atoms.

AR-ADASYN: angle radius-adaptive synthetic data generation approach for imbalanced learning

AR-ADASYN: angle radius-adaptive synthetic data generation approach for imbalanced learning

A systematic approach for data generation for intelligent fault detection and diagnosis in District Heating

This study introduces a novel systematic approach to address the challenge of labeled data scarcity for fault detection and diagnosis (FDD) in District Heating (DH) systems. To replicate real-world DH fault scenarios, we have created a controlled laboratory emulation of a generic DH substation integrated with a climate chamber. Furthermore, we present an FDD pipeline using an isolation forest and a one-class support vector machine for fault detection alongside a random forest and a support vector machine for fault diagnosis. Our research analyzed the impact of data sampling frequencies on the FDD models, revealing that shorter intervals, such as 1-min and 5-min, significantly improve FDD performance. We provide detailed information on six scenarios, including normal operation, a minor valve leak, a valve leak, a stuck valve, a high heat curve, and a temperature sensor deviation. For each scenario, we present their signature, quantifying their unique behavior and providing deeper insights into the operational implications. The signatures suggest that, while variable, faults have a consistent pattern seen in the generic DH substation. While this work contributes directly to the DH field, our methodology also extends its applicability to a broader context where labeled data is scarce.

Synthetic datasets for open software development in rare disease research

BackgroundGlobally, researchers are working on projects aiming to enhance the availability of data for rare disease research. While data sharing remains critical, developing suitable methods is challenging due to the specific sensitivity and uniqueness of rare disease data. This creates a dilemma, as there is a lack of both methods and necessary data to create appropriate approaches initially. This work contributes to bridging this gap by providing synthetic datasets that can form the foundation for such developments.MethodsUsing a hierarchical data generation approach parameterised with publicly available statistics, we generated datasets reflecting a random sample of rare disease patients from the United States (US) population. General demographics were obtained from the US Census Bureau, while information on disease prevalence, initial diagnosis, survival rates as well as race and sex ratios were obtained from the information provided by the US Centers for Disease Control and Prevention as well as the scientific literature. The software, which we have named SynthMD, was implemented in Python as open source using libraries such as Faker for generating individual data points.ResultsWe generated three datasets focusing on three specific rare diseases with broad impact on US citizens, as well as differences in affected genders and racial groups: Sickle Cell Disease, Cystic Fibrosis, and Duchenne Muscular Dystrophy. We present the statistics used to generate the datasets and study the statistical properties of output data. The datasets, as well as the code used to generate them, are available as Open Data and Open Source Software.ConclusionThe results of our work can serve as a starting point for researchers and developers working on methods and platforms that aim to improve the availability of rare disease data. Potential applications include using the datasets for testing purposes during the implementation of information systems or tailored privacy-enhancing technologies.

Benchmarking of synthetic network data: Reviewing challenges and approaches

The development of Network Intrusion Detection Systems (NIDS) requires labeled network traffic, especially to train and evaluate machine learning approaches. Besides the recording of traffic, the generation of traffic via generative models is a promising approach to obtain vast amounts of labeled data. There exist various machine learning approaches for data generation, but the assessment of the data quality is complex and not standardized. The lack of common quality criteria complicates the comparison of synthetic data generation approaches and synthetic data.Our work addresses this gap in multiple steps. Firstly, we review and categorize existing approaches for evaluating synthetic data in the network traffic domain and other data domains as well. Secondly, based on our review, we compile a setup of metrics that are suitable for the NetFlow domain, which we aggregate into two metrics Data Dissimilarity Score and Domain Dissimilarity Score. Thirdly, we evaluate the proposed metrics on real world data sets, to demonstrate their ability to distinguish between samples from different data sets. As a final step, we conduct a case study to demonstrate the application of the metrics for the evaluation of synthetic data. We calculate the metrics on samples from real NetFlow data sets to define an upper and lower bound for inter- and intra-data set similarity scores. Afterward, we generate synthetic data via Generative Adversarial Network (GAN) and Generative Pre-trained Transformer 2 (GPT-2) and apply the metrics to these synthetic data and incorporate these lower bound baseline results to obtain an objective benchmark. The application of the benchmarking process is demonstrated on three NetFlow benchmark data sets, NF-CSE-CIC-IDS2018, NF-ToN-IoT and NF-UNSW-NB15. Our demonstration indicates that this benchmark framework captures the differences in similarity between real world data and synthetic data of varying quality well, and can therefore be used to assess the quality of generated synthetic data.

Relation Extraction in Underexplored Biomedical Domains: A Diversity-optimised Sampling and Synthetic Data Generation Approach

Abstract The sparsity of labeled data is an obstacle to the development of Relation Extraction (RE) models and the completion of databases in various biomedical areas. While being of high interest in drug-discovery, the literature on natural products, reporting the identification of potential bioactive compounds from organisms, is a concrete example of such an overlooked topic. To mark the start of this new task, we created the first curated evaluation dataset and extracted literature items from the LOTUS database to build training sets. To this end, we developed a new sampler, inspired by diversity metrics in ecology, named Greedy Maximum Entropy sampler (https://github.com/idiap/gme-sampler). The strategic optimization of both balance and diversity of the selected items in the evaluation set is important given the resource-intensive nature of manual curation. After quantifying the noise in the training set, in the form of discrepancies between the text of input abstracts and the expected output labels, we explored different strategies accordingly. Framing the task as an end-to-end Relation Extraction, we evaluated the performance of standard fine-tuning (BioGPT, GPT-2, and Seq2rel) and few-shot learning with open Large Language Models (LLMs) (LLaMA 7B-65B). In addition to their evaluation in few-shot settings, we explore the potential of open LLMs as synthetic data generators and propose a new workflow for this purpose. All evaluated models exhibited substantial improvements when fine-tuned on synthetic abstracts rather than the original noisy data.We provide our best performing (F1-score = 59:0) BioGPT-Large model for end-to-end RE of natural products relationships along with all the training and evaluation datasets. See more details at https://github.com/idiap/abroad-re.

How Preceptors Support Pharmacy Learner Professional Identity Formation

ObjectivePharmacy preceptors play a role in helping learners form professional identities during experiential education. However, it is not clear what specific roles and precepting strategies best foster professional identity formation (PIF). The objective of this study was to explore how preceptors support pharmacy learner PIF. MethodsThis qualitative study used an interpretative descriptive approach. Preceptors from 5 experiential education programs were recruited using purposive sampling for individual semistructured interviews. Interviews were recorded, transcribed, coded, and analyzed by thematic analysis. Team members used a reflective and iterative approach for data analysis and generation of themes. ResultsA total of 22 participants were interviewed from various pharmacy practice settings and precept a range of learners, including introductory pharmacy practice experiences, advanced pharmacy practice experiences, and residents. Four main themes were identified to support pharmacy leaner PIF: making learners part of the practice and team, preparing learners to assume the role of a pharmacist, helping learners navigate emotions during practice experiences, and supporting learners in finding the right fit within the profession. Specific precepting strategies associated with each theme were identified. ConclusionPreceptors play an important role in supporting learners in thinking and acting as professionals while also helping navigate emotional experiences that may impact PIF and having conversations to help define learner’s future aspirations of the pharmacist they want to become. Strategies identified can inform curricular approaches and preceptor development that intentionally supports PIF.

Guided Docking as a Data Generation Approach Facilitates Structure-Based Machine Learning on Kinases.

Drug discovery pipelines nowadays rely on machine learning models to explore and evaluate large chemical spaces. While including 3D structural information is considered beneficial, structural models are hindered by the availability of protein-ligand complex structures. Exemplified for kinase drug discovery, we address this issue by generating kinase-ligand complex data using template docking for the kinase compound subset of available ChEMBL assay data. To evaluate the benefit of the created complex data, we use it to train a structure-based E(3)-invariant graph neural network. Our evaluation shows that binding affinities can be predicted with significantly higher precision by models that take synthetic binding poses into account compared to ligand- or drug-target interaction models alone.

Addressing Data Scarcity in the Medical Domain: A GPT-Based Approach for Synthetic Data Generation and Feature Extraction

This research confronts the persistent challenge of data scarcity in medical machine learning by introducing a pioneering methodology that harnesses the capabilities of Generative Pre-trained Transformers (GPT). In response to the limitations posed by a dearth of labeled medical data, our approach involves the synthetic generation of comprehensive patient discharge messages, setting a new standard in the field with GPT autonomously generating 20 fields. Through a meticulous review of the existing literature, we systematically explore GPT’s aptitude for synthetic data generation and feature extraction, providing a robust foundation for subsequent phases of the research. The empirical demonstration showcases the transformative potential of our proposed solution, presenting over 70 patient discharge messages with synthetically generated fields, including severity and chances of hospital re-admission with justification. Moreover, the data had been deployed in a mobile solution where regression algorithms autonomously identified the correlated factors for ascertaining the severity of patients’ conditions. This study not only establishes a novel and comprehensive methodology but also contributes significantly to medical machine learning, presenting the most extensive patient discharge summaries reported in the literature. The results underscore the efficacy of GPT in overcoming data scarcity challenges and pave the way for future research to refine and expand the application of GPT in diverse medical contexts.

Inverse physics–informed neural networks for digital twin–based bearing fault diagnosis under imbalanced samples

In actual engineering, insufficient bearing samples for each fault category presents a substantial obstacle to the intelligent fault diagnosis of rolling bearings. To address sample imbalance, this work explores a novel bearing fault data–generation approach based on digital twin technique. First, an inverse physics–informed neural network (PINN) is built to recognize dynamic model parameters by embedding a bearing dynamic model into a neural network. In this network, a boundary loss is designed to quickly determine the approximate ranges of parameters that can accelerate network convergence, and a true value loss is constructed for the assessment of spectral discrepancy between simulated and actual data. Then, using an inverse PINN, a bearing fault dynamic model, and real vibration data, we propose a digital twin–based fault data–generation method for producing high-quality bearing fault samples under multiple working conditions and fault modes. Finally, the developed approach is applied to generate bearing fault vibration samples under a specific working condition. The samples are used for training the diagnostic network, thus solving the issue of sample imbalance. The comparison results of several experiments suggest that the developed data-generation method effectively improves the precision of cross-working-condition bearing fault diagnosis and surpasses multiple state-of-the-art methods.

Synthetic Data Improve Survival Status Prediction Models in Early-Onset Colorectal Cancer.

In artificial intelligence-based modeling, working with a limited number of patient groups is challenging. This retrospective study aimed to evaluate whether applying synthetic data generation methods to the clinical data of small patient groups can enhance the performance of prediction models. A data set collected by the Cancer Registry Library Project from the Yonsei Cancer Center (YCC), Severance Hospital, between January 2008 and October 2020 was reviewed. Patients with colorectal cancer younger than 50 years who started their initial treatment at YCC were included. A Bayesian network-based synthesizing model was used to generate a synthetic data set, combined with the differential privacy (DP) method. A synthetic population of 5,005 was generated from a data set of 1,253 patients with 93 clinical features. The Hellinger distance and correlation difference metric were below 0.3 and 0.5, respectively, indicating no statistical difference. The overall survival by disease stage did not differ between the synthetic and original populations. Training with the synthetic data and validating with the original data showed the highest performances of 0.850, 0.836, and 0.790 for the Decision Tree, Random Forest, and XGBoost models, respectively. Comparison of synthetic data sets with different epsilon parameters from the original data sets showed improved performance >0.1%. For extremely small data sets, models using synthetic data outperformed those using only original data sets. The reidentification risk measures demonstrated that the epsilons between 0.1 and 100 fell below the baseline, indicating a preserved privacy state. The synthetic data generation approach enhances predictive modeling performance by maintaining statistical and clinical integrity, and simultaneously reduces privacy risks through the application of DP techniques.

Synthetic Data and Hierarchical Object Detection in Overhead Imagery

The performance of neural network models is often limited by the availability of big datasets. To treat this problem, we survey and develop novel synthetic data generation and augmentation techniques for enhancing low/zero-sample learning in satellite imagery. In addition to extending synthetic data generation approaches, we propose a hierarchical detection architecture to improve the utility of synthetic training samples. We consider existing techniques for producing synthetic imagery–3D models and neural style transfer–as well as introducing our own adversarially trained reskinning network, the GAN-Reskinner, to blend 3D models. Additionally, we test the value of synthetic data in a two-stage, hierarchical detection/classification model of our own construction. To test the effectiveness of synthetic imagery, we employ it in the training of detection models and our two stage model, and evaluate the resulting models on real satellite images. All modalities of synthetic data are tested extensively on practical, geospatial analysis problems. Our experiments show that synthetic data developed using our approach can often enhance detection performance, particularly when combined with some real training images. When the only source of data is synthetic, our GAN-Reskinner often boosts performance over conventionally rendered 3D models and in all cases, the hierarchical model outperforms the baseline end-to-end detection architecture.

PVS-GEN: Systematic Approach for Universal Synthetic Data Generation Involving Parameterization, Verification, and Segmentation.

Synthetic data generation addresses the challenges of obtaining extensive empirical datasets, offering benefits such as cost-effectiveness, time efficiency, and robust model development. Nonetheless, synthetic data-generation methodologies still encounter significant difficulties, including a lack of standardized metrics for modeling different data types and comparing generated results. This study introduces PVS-GEN, an automated, general-purpose process for synthetic data generation and verification. The PVS-GEN method parameterizes time-series data with minimal human intervention and verifies model construction using a specific metric derived from extracted parameters. For complex data, the process iteratively segments the empirical dataset until an extracted parameter can reproduce synthetic data that reflects the empirical characteristics, irrespective of the sensor data type. Moreover, we introduce the PoR metric to quantify the quality of the generated data by evaluating its time-series characteristics. Consequently, the proposed method can automatically generate diverse time-series data that covers a wide range of sensor types. We compared PVS-GEN with existing synthetic data-generation methodologies, and PVS-GEN demonstrated a superior performance. It generated data with a similarity of up to 37.1% across multiple data types and by 19.6% on average using the proposed metric, irrespective of the data type.

Channel Attention GAN-Based Synthetic Weed Generation for Precise Weed Identification.

Weed is a major biological factor causing declines in crop yield. However, widespread herbicide application and indiscriminate weeding with soil disturbance are of great concern because of their environmental impacts. Site-specific weed management (SSWM) refers to a weed management strategy for digital agriculture that results in low energy loss. Deep learning is crucial for developing SSWM, as it distinguishes crops from weeds and identifies weed species. However, this technique requires substantial annotated data, which necessitates expertise in weed science and agronomy. In this study, we present a channel attention mechanism-driven generative adversarial network (CA-GAN) that can generate realistic synthetic weed data. The performance of the model was evaluated using two datasets: the public segmented Plant Seedling Dataset (sPSD), featuring nine common broadleaf weeds from arable land, and the Institute for Sustainable Agro-ecosystem Services (ISAS) dataset, which includes five common summer weeds in Japan. Consequently, the synthetic dataset generated by the proposed CA-GAN obtained an 82.63% recognition accuracy on the sPSD and 93.46% on the ISAS dataset. The Fréchet inception distance (FID) score test measures the similarity between a synthetic and real dataset, and it has been shown to correlate well with human judgments of the quality of synthetic samples. The synthetic dataset achieved a low FID score (20.95 on the sPSD and 24.31 on the ISAS dataset). Overall, the experimental results demonstrated that the proposed method outperformed previous state-of-the-art GAN models in terms of image quality, diversity, and discriminability, making it a promising approach for synthetic agricultural data generation.

Unlocking biomedical data sharing: A structured approach with digital twins and artificial intelligence (AI) for open health sciences.

Data sharing promotes the scientific progress. However, not all data can be shared freely due to privacy issues. This work is intended to foster FAIR sharing of sensitive data exemplary in the biomedical domain, via an integrated computational approach for utilizing and enriching individual datasets by scientists without coding experience. We present an in silico pipeline for openly sharing controlled materials by generating synthetic data. Additionally, it addresses the issue of inexperience to computational methods in a non-IT-affine domain by making use of a cyberinfrastructure that runs and enables sharing of computational notebooks without the need of local software installation. The use of a digital twin based on cancer datasets serves as exemplary use case for making biomedical data openly available. Quantitative and qualitative validation of model output as well as a study on user experience are conducted. The metadata approach describes generalizable descriptors for computational models, and outlines how to profit from existing data resources for validating computational models. The use of a virtual lab book cooperatively developed using a cloud-based data management and analysis system functions as showcase enabling easy interaction between users. Qualitative testing revealed a necessity for comprehensive guidelines furthering acceptance by various users. The introduced framework presents an integrated approach for data generation and interpolating incomplete data, promoting Open Science through reproducibility of results and methods. The system can be expanded from the biomedical to any other domain while future studies integrating an enhanced graphical user interface could increase interdisciplinary applicability.

KI-MAG: A knowledge-infused abstractive question answering system in medical domain

Abstractive question-answering (QA) has emerged as a prominent area in Natural Language Processing (NLP) due to its ability to produce concise and human-like responses, particularly with the advancement of Large Language Models. Despite its potential, abstractive QA suffers from challenges like the need for extensive training data and the generation of incorrect entities and out-of-context words in the responses. In safety–critical domains like medical and clinical settings, such issues are unacceptable and may compromise the accuracy and reliability of generated answers. We proposed KI-MAG (Knowledge-Infused Medical Abstractive Generator) model, a novel Knowledge-Infused Abstractive Question Answering System specifically designed for the medical domain. KI-MAG aims to address the aforementioned limitations and enhance the correctness of generated responses while mitigating data sparsity concerns. The KI-MAG system produces more precise and informative answers by incorporating relevant medical entities into the model’s generation process. Furthermore, we adopt a synthetic data generation approach using question–answer pairs to overcome the challenge of limited training data in the medical domain. These synthetic pairs augment the original dataset, resulting in better model generalization and improved performance. Our extensive experimental evaluations demonstrate the effectiveness of the KI-MAG system. Compared to traditional abstractive QA models, our approach exhibits a substantial increase of approximately 15% in Blue-1, Blue-2, Blue-3, and Blue-4 scores, indicating a remarkable improvement in answer accuracy and overall quality of responses. Overall, our Knowledge-Infused Abstractive Question Answering System in the Medical Domain (KI-MAG) presents a promising solution to enhance the performance and reliability of abstractive QA models in safety–critical medical applications where precision and correctness of answers are of utmost importance.

GAMMA: Graph Attention Model for Multiple Agents to Solve Team Orienteering Problem With Multiple Depots.

In this work, we present an attention-based encoder-decoder model to approximately solve the team orienteering problem with multiple depots (TOPMD). The TOPMD instance is an NP-hard combinatorial optimization problem that involves multiple agents (or autonomous vehicles) and not purely Euclidean (straight line distance) graph edge weights. In addition, to avoid tedious computations on dataset creation, we provide an approach to generate synthetic data on the fly for effectively training the model. Furthermore, to evaluate our proposed model, we conduct two experimental studies on the multi-agent reconnaissance mission planning problem formulated as TOPMD. First, we characterize the model based on the training configurations to understand the scalability of the proposed approach to unseen configurations. Second, we evaluate the solution quality of the model against several baselines-heuristics, competing machine learning (ML), and exact approaches, on several reconnaissance scenarios. The experimental results indicate that training the model with a maximum number of agents, a moderate number of targets (or nodes to visit), and moderate travel length, performs well across a variety of conditions. Furthermore, the results also reveal that the proposed approach offers a more tractable and higher quality (or competitive) solution in comparison with existing attention-based models, stochastic heuristic approach, and standard mixed-integer programming solver under the given experimental conditions. Finally, the different experimental evaluations reveal that the proposed data generation approach for training the model is highly effective.