Maria Ganopoulou, Efstratios Kontopoulos, Konstantinos Fokianos, Dimitris Koparanis, L. Angelis, Ioannis Kotsianidis, Theodoros Moysiadis
Questionnaires on health-related quality of life (HRQoL) play a crucial role in managing patients by revealing insights into physical, psychological, lifestyle, and social factors affecting well-being. A methodological aspect that has not been adequately explored yet, and is of considerable potential, is causal discovery. This study explored causal discovery techniques within HRQoL, assessed various considerations for reliable estimation, and proposed means for interpreting outcomes. Five causal structure learning algorithms were employed to examine different aspects in structure estimation based on simulated data derived from HRQoL-related directed acyclic graphs. The performance of the algorithms was assessed based on various measures related to the differences between the true and estimated structures. Moreover, the Resource Description Framework was adopted to represent the responses to the HRQoL questionnaires and the detected cause–effect relationships among the questions, resulting in semantic knowledge graphs which are structured representations of interconnected information. It was found that the structure estimation was impacted negatively by the structure’s complexity and favorably by increasing the sample size. The performance of the algorithms over increasing sample size exhibited a similar pattern, with distinct differences being observed for small samples. This study illustrates the dynamics of causal discovery in HRQoL-related research, highlights aspects that should be addressed in estimation, and fosters the shareability and interoperability of the output based on globally established standards. Thus, it provides critical insights in this context, further promoting the critical role of HRQoL questionnaires in advancing patient-centered care and management.
{"title":"Delving into Causal Discovery in Health-Related Quality of Life Questionnaires","authors":"Maria Ganopoulou, Efstratios Kontopoulos, Konstantinos Fokianos, Dimitris Koparanis, L. Angelis, Ioannis Kotsianidis, Theodoros Moysiadis","doi":"10.3390/a17040138","DOIUrl":"https://doi.org/10.3390/a17040138","url":null,"abstract":"Questionnaires on health-related quality of life (HRQoL) play a crucial role in managing patients by revealing insights into physical, psychological, lifestyle, and social factors affecting well-being. A methodological aspect that has not been adequately explored yet, and is of considerable potential, is causal discovery. This study explored causal discovery techniques within HRQoL, assessed various considerations for reliable estimation, and proposed means for interpreting outcomes. Five causal structure learning algorithms were employed to examine different aspects in structure estimation based on simulated data derived from HRQoL-related directed acyclic graphs. The performance of the algorithms was assessed based on various measures related to the differences between the true and estimated structures. Moreover, the Resource Description Framework was adopted to represent the responses to the HRQoL questionnaires and the detected cause–effect relationships among the questions, resulting in semantic knowledge graphs which are structured representations of interconnected information. It was found that the structure estimation was impacted negatively by the structure’s complexity and favorably by increasing the sample size. The performance of the algorithms over increasing sample size exhibited a similar pattern, with distinct differences being observed for small samples. This study illustrates the dynamics of causal discovery in HRQoL-related research, highlights aspects that should be addressed in estimation, and fosters the shareability and interoperability of the output based on globally established standards. Thus, it provides critical insights in this context, further promoting the critical role of HRQoL questionnaires in advancing patient-centered care and management.","PeriodicalId":7636,"journal":{"name":"Algorithms","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140376765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Drones play a pivotal role in various industries of Industry 4.0. For achieving the application of drones in a dynamic environment, finding a clear path for their autonomous flight requires more research. This paper addresses the problem of finding a navigation path for an autonomous drone based on visual scene information. A deep learning-based object detection approach can localize obstacles detected in a scene. Considering this approach, we propose a solution framework that includes masking with a color-based segmentation method to identify an empty area where the drone can fly. The scene is described using segmented regions and localization points. The proposed approach can be used to remotely guide drones in dynamic environments that have poor coverage from global positioning systems. The simulation results show that the proposed framework with object detection and the proposed masking technique support drone navigation in a dynamic environment based only on the visual input from the front field of view.
{"title":"Testing a Vision-Based Autonomous Drone Navigation Model in a Forest Environment","authors":"Alvin Lee, Suet-Peng Yong, W. Pedrycz, J. Watada","doi":"10.3390/a17040139","DOIUrl":"https://doi.org/10.3390/a17040139","url":null,"abstract":"Drones play a pivotal role in various industries of Industry 4.0. For achieving the application of drones in a dynamic environment, finding a clear path for their autonomous flight requires more research. This paper addresses the problem of finding a navigation path for an autonomous drone based on visual scene information. A deep learning-based object detection approach can localize obstacles detected in a scene. Considering this approach, we propose a solution framework that includes masking with a color-based segmentation method to identify an empty area where the drone can fly. The scene is described using segmented regions and localization points. The proposed approach can be used to remotely guide drones in dynamic environments that have poor coverage from global positioning systems. The simulation results show that the proposed framework with object detection and the proposed masking technique support drone navigation in a dynamic environment based only on the visual input from the front field of view.","PeriodicalId":7636,"journal":{"name":"Algorithms","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140374094","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Back pain is regularly associated with biomechanical changes in the spine. The traditional methods to assess spine biomechanics use ionising radiation. Vertebral Metrics (VM) is a non-invasive instrument developed by the authors in previous research that assesses the spinous processes’ position. However, the spine model used by VM is not accurate. To overcome it, the present paper proposes a pioneering and simple articulated model of the spine built through the data collected by VM. The model is based on the spring–mass system and uses the Levenberg–Marquardt algorithm to find the arrangement of vertebral bodies. It represents the spine as rigid geometric transformations from one vertebra to the other when the extremity vertebrae are stationary. The validation process used the Bland–Altman method to compare the Cobb and the intervertebral angles computed by the model with the radiographic exams of eight patients diagnosed with Ankylosing Spondylitis. The results suggest that the model is valid; however, previous clinical information would improve outcomes by customising the lower and upper vertebrae positions, since the study revealed that the C6 rotation slightly influences the computed angles. Applying VM with the new model could make a difference in preventing, monitoring, and early diagnosing spinal disorders.
{"title":"An Innovative Mathematical Model of the Spine: Predicting Cobb and Intervertebral Angles Using the 3D Position of the Spinous Processes Measured by Vertebral Metrics","authors":"A. Gabriel, Cláudia Quaresma, P. Vieira","doi":"10.3390/a17040134","DOIUrl":"https://doi.org/10.3390/a17040134","url":null,"abstract":"Back pain is regularly associated with biomechanical changes in the spine. The traditional methods to assess spine biomechanics use ionising radiation. Vertebral Metrics (VM) is a non-invasive instrument developed by the authors in previous research that assesses the spinous processes’ position. However, the spine model used by VM is not accurate. To overcome it, the present paper proposes a pioneering and simple articulated model of the spine built through the data collected by VM. The model is based on the spring–mass system and uses the Levenberg–Marquardt algorithm to find the arrangement of vertebral bodies. It represents the spine as rigid geometric transformations from one vertebra to the other when the extremity vertebrae are stationary. The validation process used the Bland–Altman method to compare the Cobb and the intervertebral angles computed by the model with the radiographic exams of eight patients diagnosed with Ankylosing Spondylitis. The results suggest that the model is valid; however, previous clinical information would improve outcomes by customising the lower and upper vertebrae positions, since the study revealed that the C6 rotation slightly influences the computed angles. Applying VM with the new model could make a difference in preventing, monitoring, and early diagnosing spinal disorders.","PeriodicalId":7636,"journal":{"name":"Algorithms","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140384843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
J. Romero-González, Diana-Margarita Córdova-Esparza, Juan R. Terven, A. Herrera-Navarro, Hugo Jiménez-Hernández
This paper introduces a novel background subtraction method that utilizes texture-level analysis based on the Gabor filter bank and statistical moments. The method addresses the challenge of accurately detecting moving objects that exhibit similar color intensity variability or texture to the surrounding environment, which conventional methods struggle to handle effectively. The proposed method accurately distinguishes between foreground and background objects by capturing different frequency components using the Gabor filter bank and quantifying the texture level through statistical moments. Extensive experimental evaluations use datasets featuring varying lighting conditions, uniform and non-uniform textures, shadows, and dynamic backgrounds. The performance of the proposed method is compared against other existing methods using metrics such as sensitivity, specificity, and false positive rate. The experimental results demonstrate that the proposed method outperforms other methods in accuracy and robustness. It effectively handles scenarios with complex backgrounds, lighting changes, and objects that exhibit similar texture or color intensity as the background. Our method retains object structure while minimizing false detections and noise. This paper provides valuable insights into computer vision and object detection, offering a promising solution for accurate foreground detection in various applications such as video surveillance and motion tracking.
{"title":"Background Subtraction for Dynamic Scenes Using Gabor Filter Bank and Statistical Moments","authors":"J. Romero-González, Diana-Margarita Córdova-Esparza, Juan R. Terven, A. Herrera-Navarro, Hugo Jiménez-Hernández","doi":"10.3390/a17040133","DOIUrl":"https://doi.org/10.3390/a17040133","url":null,"abstract":"This paper introduces a novel background subtraction method that utilizes texture-level analysis based on the Gabor filter bank and statistical moments. The method addresses the challenge of accurately detecting moving objects that exhibit similar color intensity variability or texture to the surrounding environment, which conventional methods struggle to handle effectively. The proposed method accurately distinguishes between foreground and background objects by capturing different frequency components using the Gabor filter bank and quantifying the texture level through statistical moments. Extensive experimental evaluations use datasets featuring varying lighting conditions, uniform and non-uniform textures, shadows, and dynamic backgrounds. The performance of the proposed method is compared against other existing methods using metrics such as sensitivity, specificity, and false positive rate. The experimental results demonstrate that the proposed method outperforms other methods in accuracy and robustness. It effectively handles scenarios with complex backgrounds, lighting changes, and objects that exhibit similar texture or color intensity as the background. Our method retains object structure while minimizing false detections and noise. This paper provides valuable insights into computer vision and object detection, offering a promising solution for accurate foreground detection in various applications such as video surveillance and motion tracking.","PeriodicalId":7636,"journal":{"name":"Algorithms","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140384908","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Salient object detection (SOD) aims to identify the most visually striking objects in a scene, simulating the function of the biological visual attention system. The attention mechanism in deep learning is commonly used as an enhancement strategy which enables the neural network to concentrate on the relevant parts when processing input data, effectively improving the model’s learning and prediction abilities. Existing saliency object detection methods based on RGB deep learning typically treat all regions equally by using the extracted features, overlooking the fact that different regions have varying contributions to the final predictions. Based on the U2Net algorithm, this paper incorporates the split coordinate channel attention (SCCA) mechanism into the feature extraction stage. SCCA conducts spatial transformation in width and height dimensions to efficiently extract the location information of the target to be detected. While pixel-level semantic segmentation based on annotation has been successful, it assigns the same weight to each pixel which leads to poor performance in detecting the boundary of objects. In this paper, the Canny edge detection loss is incorporated into the loss calculation stage to improve the model’s ability to detect object edges. Based on the DUTS and HKU-IS datasets, experiments confirm that the proposed strategies effectively enhance the model’s detection performance, resulting in a 0.8% and 0.7% increase in the F1-score of U2Net. This paper also compares the traditional attention modules with the newly proposed attention, and the SCCA attention module achieves a top-three performance in prediction time, mean absolute error (MAE), F1-score, and model size on both experimental datasets.
{"title":"Application of Split Coordinate Channel Attention Embedding U2Net in Salient Object Detection","authors":"Yuhuan Wu, Yonghong Wu","doi":"10.3390/a17030109","DOIUrl":"https://doi.org/10.3390/a17030109","url":null,"abstract":"Salient object detection (SOD) aims to identify the most visually striking objects in a scene, simulating the function of the biological visual attention system. The attention mechanism in deep learning is commonly used as an enhancement strategy which enables the neural network to concentrate on the relevant parts when processing input data, effectively improving the model’s learning and prediction abilities. Existing saliency object detection methods based on RGB deep learning typically treat all regions equally by using the extracted features, overlooking the fact that different regions have varying contributions to the final predictions. Based on the U2Net algorithm, this paper incorporates the split coordinate channel attention (SCCA) mechanism into the feature extraction stage. SCCA conducts spatial transformation in width and height dimensions to efficiently extract the location information of the target to be detected. While pixel-level semantic segmentation based on annotation has been successful, it assigns the same weight to each pixel which leads to poor performance in detecting the boundary of objects. In this paper, the Canny edge detection loss is incorporated into the loss calculation stage to improve the model’s ability to detect object edges. Based on the DUTS and HKU-IS datasets, experiments confirm that the proposed strategies effectively enhance the model’s detection performance, resulting in a 0.8% and 0.7% increase in the F1-score of U2Net. This paper also compares the traditional attention modules with the newly proposed attention, and the SCCA attention module achieves a top-three performance in prediction time, mean absolute error (MAE), F1-score, and model size on both experimental datasets.","PeriodicalId":7636,"journal":{"name":"Algorithms","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140078268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Anton Kolosnitsyn, Oleg Khamisov, Eugene Semenkin, Vladimir Nelyub
We consider the Golden Section and Parabola Methods for solving univariate optimization problems. For multivariate problems, we use these methods as line search procedures in combination with well-known zero-order methods such as the coordinate descent method, the Hooke and Jeeves method, and the Rosenbrock method. A comprehensive numerical comparison of the obtained versions of zero-order methods is given in the present work. The set of test problems includes nonconvex functions with a large number of local and global optimum points. Zero-order methods combined with the Parabola method demonstrate high performance and quite frequently find the global optimum even for large problems (up to 100 variables).
{"title":"Application of the Parabola Method in Nonconvex Optimization","authors":"Anton Kolosnitsyn, Oleg Khamisov, Eugene Semenkin, Vladimir Nelyub","doi":"10.3390/a17030107","DOIUrl":"https://doi.org/10.3390/a17030107","url":null,"abstract":"We consider the Golden Section and Parabola Methods for solving univariate optimization problems. For multivariate problems, we use these methods as line search procedures in combination with well-known zero-order methods such as the coordinate descent method, the Hooke and Jeeves method, and the Rosenbrock method. A comprehensive numerical comparison of the obtained versions of zero-order methods is given in the present work. The set of test problems includes nonconvex functions with a large number of local and global optimum points. Zero-order methods combined with the Parabola method demonstrate high performance and quite frequently find the global optimum even for large problems (up to 100 variables).","PeriodicalId":7636,"journal":{"name":"Algorithms","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140085051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Somayeh Shahrabadi, T. Adão, Emanuel Peres, R. Morais, L. G. Magalhães, Victor Alves
The proliferation of classification-capable artificial intelligence (AI) across a wide range of domains (e.g., agriculture, construction, etc.) has been allowed to optimize and complement several tasks, typically operationalized by humans. The computational training that allows providing such support is frequently hindered by various challenges related to datasets, including the scarcity of examples and imbalanced class distributions, which have detrimental effects on the production of accurate models. For a proper approach to these challenges, strategies smarter than the traditional brute force-based K-fold cross-validation or the naivety of hold-out are required, with the following main goals in mind: (1) carrying out one-shot, close-to-optimal data arrangements, accelerating conventional training optimization; and (2) aiming at maximizing the capacity of inference models to its fullest extent while relieving computational burden. To that end, in this paper, two image-based feature-aware dataset splitting approaches are proposed, hypothesizing a contribution towards attaining classification models that are closer to their full inference potential. Both rely on strategic image harvesting: while one of them hinges on weighted random selection out of a feature-based clusters set, the other involves a balanced picking process from a sorted list that stores data features’ distances to the centroid of a whole feature space. Comparative tests on datasets related to grapevine leaves phenotyping and bridge defects showcase promising results, highlighting a viable alternative to K-fold cross-validation and hold-out methods.
具有分类能力的人工智能（AI）在广泛领域（如农业、建筑等）的普及，使得通常由人类操作的若干任务得以优化和补充。提供此类支持的计算训练经常受到与数据集有关的各种挑战的阻碍，包括示例稀缺和类分布不平衡，这对准确模型的生成产生了不利影响。要正确应对这些挑战，就需要比传统的基于蛮力的 K 折交叉验证或天真无邪的搁置策略更聪明的策略，主要目标如下：（1）进行一次性、接近最优的数据安排，加速传统的训练优化；（2）旨在最大限度地提高推理模型的能力，同时减轻计算负担。为此，本文提出了两种基于图像的特征感知数据集拆分方法，希望能为实现更接近其全部推理潜力的分类模型做出贡献。这两种方法都依赖于策略性的图像采集：其中一种方法依赖于从基于特征的聚类集中进行加权随机选择，而另一种方法则涉及从存储数据特征与整个特征空间中心点距离的排序列表中进行均衡挑选的过程。在与葡萄叶片表型和桥梁缺陷相关的数据集上进行的比较测试显示了良好的结果，突出了 K 折交叉验证和保留方法的可行替代方案。
{"title":"Automatic Optimization of Deep Learning Training through Feature-Aware-Based Dataset Splitting","authors":"Somayeh Shahrabadi, T. Adão, Emanuel Peres, R. Morais, L. G. Magalhães, Victor Alves","doi":"10.3390/a17030106","DOIUrl":"https://doi.org/10.3390/a17030106","url":null,"abstract":"The proliferation of classification-capable artificial intelligence (AI) across a wide range of domains (e.g., agriculture, construction, etc.) has been allowed to optimize and complement several tasks, typically operationalized by humans. The computational training that allows providing such support is frequently hindered by various challenges related to datasets, including the scarcity of examples and imbalanced class distributions, which have detrimental effects on the production of accurate models. For a proper approach to these challenges, strategies smarter than the traditional brute force-based K-fold cross-validation or the naivety of hold-out are required, with the following main goals in mind: (1) carrying out one-shot, close-to-optimal data arrangements, accelerating conventional training optimization; and (2) aiming at maximizing the capacity of inference models to its fullest extent while relieving computational burden. To that end, in this paper, two image-based feature-aware dataset splitting approaches are proposed, hypothesizing a contribution towards attaining classification models that are closer to their full inference potential. Both rely on strategic image harvesting: while one of them hinges on weighted random selection out of a feature-based clusters set, the other involves a balanced picking process from a sorted list that stores data features’ distances to the centroid of a whole feature space. Comparative tests on datasets related to grapevine leaves phenotyping and bridge defects showcase promising results, highlighting a viable alternative to K-fold cross-validation and hold-out methods.","PeriodicalId":7636,"journal":{"name":"Algorithms","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-02-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140416529","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In an era where technological advancements are rapidly transforming industries, healthcare is the primary beneficiary of such progress [...]
在技术进步迅速改变各行各业的时代，医疗保健是这种进步的主要受益者 [...]
{"title":"Artificial Intelligence Algorithms for Healthcare","authors":"D. Chumachenko, Sergiy Yakovlev","doi":"10.3390/a17030105","DOIUrl":"https://doi.org/10.3390/a17030105","url":null,"abstract":"In an era where technological advancements are rapidly transforming industries, healthcare is the primary beneficiary of such progress [...]","PeriodicalId":7636,"journal":{"name":"Algorithms","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140421988","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Varsha S. Lalapura, Veerender Reddy Bhimavarapu, J. Amudha, H. Satheesh
The Recurrent Neural Networks (RNNs) are an essential class of supervised learning algorithms. Complex tasks like speech recognition, machine translation, sentiment classification, weather prediction, etc., are now performed by well-trained RNNs. Local or cloud-based GPU machines are used to train them. However, inference is now shifting to miniature, mobile, IoT devices and even micro-controllers. Due to their colossal memory and computing requirements, mapping RNNs directly onto resource-constrained platforms is arcane and challenging. The efficacy of edge-intelligent RNNs (EI-RNNs) must satisfy both performance and memory-fitting requirements at the same time without compromising one for the other. This study’s aim was to provide an empirical evaluation and optimization of historic as well as recent RNN architectures for high-performance and low-memory footprint goals. We focused on Human Activity Recognition (HAR) tasks based on wearable sensor data for embedded healthcare applications. We evaluated and optimized six different recurrent units, namely Vanilla RNNs, Long Short-Term Memory (LSTM) units, Gated Recurrent Units (GRUs), Fast Gated Recurrent Neural Networks (FGRNNs), Fast Recurrent Neural Networks (FRNNs), and Unitary Gated Recurrent Neural Networks (UGRNNs) on eight publicly available time-series HAR datasets. We used the hold-out and cross-validation protocols for training the RNNs. We used low-rank parameterization, iterative hard thresholding, and spare retraining compression for RNNs. We found that efficient training (i.e., dataset handling and preprocessing procedures, hyperparameter tuning, and so on, and suitable compression methods (like low-rank parameterization and iterative pruning) are critical in optimizing RNNs for performance and memory efficiency. We implemented the inference of the optimized models on Raspberry Pi.
{"title":"A Systematic Evaluation of Recurrent Neural Network Models for Edge Intelligence and Human Activity Recognition Applications","authors":"Varsha S. Lalapura, Veerender Reddy Bhimavarapu, J. Amudha, H. Satheesh","doi":"10.3390/a17030104","DOIUrl":"https://doi.org/10.3390/a17030104","url":null,"abstract":"The Recurrent Neural Networks (RNNs) are an essential class of supervised learning algorithms. Complex tasks like speech recognition, machine translation, sentiment classification, weather prediction, etc., are now performed by well-trained RNNs. Local or cloud-based GPU machines are used to train them. However, inference is now shifting to miniature, mobile, IoT devices and even micro-controllers. Due to their colossal memory and computing requirements, mapping RNNs directly onto resource-constrained platforms is arcane and challenging. The efficacy of edge-intelligent RNNs (EI-RNNs) must satisfy both performance and memory-fitting requirements at the same time without compromising one for the other. This study’s aim was to provide an empirical evaluation and optimization of historic as well as recent RNN architectures for high-performance and low-memory footprint goals. We focused on Human Activity Recognition (HAR) tasks based on wearable sensor data for embedded healthcare applications. We evaluated and optimized six different recurrent units, namely Vanilla RNNs, Long Short-Term Memory (LSTM) units, Gated Recurrent Units (GRUs), Fast Gated Recurrent Neural Networks (FGRNNs), Fast Recurrent Neural Networks (FRNNs), and Unitary Gated Recurrent Neural Networks (UGRNNs) on eight publicly available time-series HAR datasets. We used the hold-out and cross-validation protocols for training the RNNs. We used low-rank parameterization, iterative hard thresholding, and spare retraining compression for RNNs. We found that efficient training (i.e., dataset handling and preprocessing procedures, hyperparameter tuning, and so on, and suitable compression methods (like low-rank parameterization and iterative pruning) are critical in optimizing RNNs for performance and memory efficiency. We implemented the inference of the optimized models on Raspberry Pi.","PeriodicalId":7636,"journal":{"name":"Algorithms","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140420904","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fengwei Jing, Fenghe Li, Yong Song, Jie Li, Zhanbiao Feng, Jin Guo
The concept of production stability in hot strip rolling encapsulates the ability of a production line to consistently maintain its output levels and uphold the quality of its products, thus embodying the steady and uninterrupted nature of the production yield. This scholarly paper focuses on the paramount looper equipment in the finishing rolling area, utilizing it as a case study to investigate approaches for identifying the origins of instabilities, specifically when faced with inadequate looper performance. Initially, the paper establishes the equipment process accuracy evaluation (EPAE) model for the looper, grounded in the precision of the looper’s operational process, to accurately depict the looper’s functioning state. Subsequently, it delves into the interplay between the EPAE metrics and overall production stability, advocating for the use of EPAE scores as direct indicators of production stability. The study further introduces a novel algorithm designed to trace the root causes of issues, categorizing them into material, equipment, and control factors, thereby facilitating on-site fault rectification. Finally, the practicality and effectiveness of this methodology are substantiated through its application on the 2250 hot rolling equipment production line. This paper provides a new approach for fault tracing in the hot rolling process.
{"title":"Root Cause Tracing Using Equipment Process Accuracy Evaluation for Looper in Hot Rolling","authors":"Fengwei Jing, Fenghe Li, Yong Song, Jie Li, Zhanbiao Feng, Jin Guo ","doi":"10.3390/a17030102","DOIUrl":"https://doi.org/10.3390/a17030102","url":null,"abstract":"The concept of production stability in hot strip rolling encapsulates the ability of a production line to consistently maintain its output levels and uphold the quality of its products, thus embodying the steady and uninterrupted nature of the production yield. This scholarly paper focuses on the paramount looper equipment in the finishing rolling area, utilizing it as a case study to investigate approaches for identifying the origins of instabilities, specifically when faced with inadequate looper performance. Initially, the paper establishes the equipment process accuracy evaluation (EPAE) model for the looper, grounded in the precision of the looper’s operational process, to accurately depict the looper’s functioning state. Subsequently, it delves into the interplay between the EPAE metrics and overall production stability, advocating for the use of EPAE scores as direct indicators of production stability. The study further introduces a novel algorithm designed to trace the root causes of issues, categorizing them into material, equipment, and control factors, thereby facilitating on-site fault rectification. Finally, the practicality and effectiveness of this methodology are substantiated through its application on the 2250 hot rolling equipment production line. This paper provides a new approach for fault tracing in the hot rolling process.","PeriodicalId":7636,"journal":{"name":"Algorithms","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140430831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：