Training Costs Research Articles

A-SOiD, an active-learning platform for expert-guided, data-efficient discovery of behavior.

To identify and extract naturalistic behavior, two methods have become popular: supervised and unsupervised. Each approach carries its own strengths and weaknesses (for example, user bias, training cost, complexity and action discovery), which the user must consider in their decision. Here, an active-learning platform, A-SOiD, blends these strengths, and in doing so, overcomes several of their inherent drawbacks. A-SOiD iteratively learns user-defined groups with a fraction of the usual training data, while attaining expansive classification through directed unsupervised classification. In socially interacting mice, A-SOiD outperformed standard methods despite requiring 85% less training data. Additionally, it isolated ethologically distinct mouse interactions via unsupervised classification. We observed similar performance and efficiency using nonhuman primate and human three-dimensional pose data. In both cases, the transparency in A-SOiD's cluster definitions revealed the defining features of the supervised classification through a game-theoretic approach. To facilitate use, A-SOiD comes as an intuitive, open-source interface for efficient segmentation of user-defined behaviors and discovered sub-actions.

Comparing Medical Training Costs Internationally: A Systematic Review.

Acquiring a thorough understanding of the expenses linked to the education of health sciences students is crucial for effective university planning, budgeting, and overall preparedness. This systematic review aimed to identify and compare the per capita costs associated with educating medical and other health science students internationally-particularly emphasizing the context of Iran. A systematic review was conducted in 2023 according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. The search covered the period from January 1, 2000, to November 11, 2022, using MeSH and EMTREE terms. Databases such as PubMed, Scopus, Web of Science, Embase, and Iranian databases were searched. Manual searches were performed using Google and Google Scholar. The study retrieved 1336 publications from bibliometric databases and, following thorough screening, included 8 relevant articles from 5 countries (Australia, Iran, United States, Thailand, and Vietnam). An additional 17 relevant articles from Iranian databases were also included. Based on USD purchasing power parity (PPP) 2019, the results show that the mean per capita cost of training a medical student for 1 academic year in Iran is $61,493.86 (range, $28,102-$133,603; standard deviation, $35,476.03). In comparison, the cost of training a medical student for 1 year is $263,305 in the United States and $44,674 in Australia. In Thailand and Vietnam, a 6-year medical program costs is $284,058 and $69,323, respectively. Moreover, according to most studies, training students in other health sciences in Iran for 1 academic year generally costs ˂$20,000 (PPP 2019). The study reveals that the costs associated with medical student education in Iran exceed that of most countries, second only to the United States. These findings highlight the importance of such data in improving the efficiency, sustainability, and informed resource allocation of global medical education programs for future planning and budgeting.

Distributed Feature Selection for Power System Dynamic Security Region Based on Grid-Partition and Fuzzy-Rough Sets

In order to satisfy the requirements of modern online security assessment of power systems with continuously increasing complexity in terms of structure and scale, it is desirable to develop a power system dynamic security region (DSR) analysis. However, data-driven methods suffer from expensive model training costs and overfitting when determining DSR boundaries with high-dimensional grid features. Given this problem, a distributed feature selection method based on grid partition and fuzzy-rough sets is proposed in this paper. The method first employs the Louvain algorithm to partition the power grid and divide the original feature set so that high-dimensional features can be allocated to multiple computational units for distributed screening. At this point, the connections between features of different computational units are minimized to a relatively low level, thereby avoiding large errors in the distributed results. Then, an incremental search algorithm based on the fuzzy-rough set theory (FRST) is used for feature selection at each computational unit, which can effectively take into account the intrinsic connections between features. Finally, the results of all computational units are integrated in the coordination unit to complete the overall feature selection. The experimental results based on the IEEE-39 bus system show that the proposed method can help simplify the power system DSR analysis with high-dimensional features by screening the critical features. And compared with other commonly used filter methods, it has higher screening accuracy and lower time costs.

Prediction of Training Cost and Difficulty for Aircraft-Type Transition Based on Similarity Assessment

As aviation technology advances, numerous new aircraft enter the market. These not only offer airlines technological and fuel efficiency advantages but also present the challenge of how to conduct pilots’ aircraft-type transition training efficiently and economically. To address this issue, this study designed a methodology to quantitatively assess the similarity in panel display control design and standard operating procedures (SOPs) between aircraft types. Then, by combining the results of a questionnaire survey on A320, A330, B737, and B777 transition training and training cost data, it was verified quantitatively that inter-aircraft similarity has a positive impact on reducing the difficulty and cost of transition training. Taking the similarity in aircraft types as a feature, the KNN algorithm was used to successfully construct a difficulty prediction model for the training program of aircraft-type transition training. To overcome the limitation of insufficient training cost data volume, this study adopts the transfer learning method to construct a prediction model of the transition training cost, and the final significant prediction accuracy proves the effectiveness of the method. The research in this paper not only provides strong data support for the resource planning and cost management of airlines’ aircraft-type transition training but also provides new research perspectives and methodological guidance for the field of aviation training.

Lightweight land cover classification via semantic segmentation of remote sensing imagery and analysis of influencing factors

Land cover classification is of great value and can be widely used in many fields. Earlier land cover classification methods used traditional image segmentation techniques, which cannot fully and comprehensively extract the ground information in remote sensing images. Therefore, it is necessary to integrate the advanced techniques of deep learning into the study of semantic segmentation of remote sensing images. However, most of current high-resolution image segmentation networks have disadvantages such as large parameters and high network training cost. In view of the problems above, a lightweight land cover classification model via semantic segmentation, DeepGDLE, is proposed in this paper. The model DeepGDLE is designed on the basis of DeeplabV3+ network and utilizes the GhostNet network instead of the backbone feature extraction network in the encoder. Using Depthwise Separable Convolution (DSC) instead of dilation convolution. This reduces the number of parameters and increases the computational speed of the model. By optimizing the dilation rate of parallel convolution in the ASPP module, the “grid effect” is avoided. ECANet lightweight channel attention mechanism is added after the feature extraction module and the pyramid pooling module to focus on the important weights of the model. Finally, the loss function Focal Loss is utilized to solve the problem of category imbalance in the dataset. As a result, the model DeepGDLE effectively reduces the parameters of the network model and the network training cost. And extensive experiments compared with several existing semantic segmentation algorithms such as DeeplabV3+, UNet, SegNet, etc. show that DeepGDLE improves the quality and efficiency of image segmentation. Therefore, compared to other networks, the DeepGDLE network model can be more effectively applied to land cover classification. In addition, in order to investigate the effects of different factors on the semantic segmentation performance of remote sensing images and to verify the robustness of the DeepGDLE model, a new remote sensing image dataset, FRSID, is constructed in this paper. This dataset takes into account more influences than the public dataset. The experimental results show that on the WHDLD dataset, the experimental metrics mIoU, mPA, and mRecall of the proposed model, DeepGDLE, are 62.29%, 72.85%, and 72.46%, respectively. On the FRSID dataset, the metrics mIoU, mPA, and mRecall are 65.89%, 74.43%, and 74.08%, respectively. For the future scope of research in this field, it may focus on the fusion of multi-source remote sensing data and the intelligent interpretation of remote sensing images.

Gene selection for high dimensional biological datasets using hybrid island binary artificial bee colony with chaos game optimization

Microarray technology, as applied to the fields of bioinformatics, biotechnology, and bioengineering, has made remarkable progress in both the treatment and prediction of many biological problems. However, this technology presents a critical challenge due to the size of the numerous genes present in the high-dimensional biological datasets associated with an experiment, which leads to a curse of dimensionality on biological data. Such high dimensionality of real biological data sets not only increases memory requirements and training costs, but also reduces the ability of learning algorithms to generalise. Consequently, multiple feature selection (FS) methods have been proposed by researchers to choose the most significant and precise subset of classified genes from gene expression datasets while maintaining high classification accuracy. In this research work, a novel binary method called iBABC-CGO based on the island model of the artificial bee colony algorithm, combined with the chaos game optimization algorithm and SVM classifier, is suggested for FS problems using gene expression data. Due to the binary nature of FS problems, two distinct transfer functions are employed for converting the continuous search space into a binary one, thus improving the efficiency of the exploration and exploitation phases. The suggested strategy is tested on a variety of biological datasets with different scales and compared to popular metaheuristic-based, filter-based, and hybrid FS methods. Experimental results supplemented with the statistical measures, box plots, Wilcoxon tests, Friedman tests, and radar plots demonstrate that compared to prior methods, the proposed iBABC-CGO exhibit competitive performance in terms of classification accuracy, selection of the most relevant subset of genes, data variability, and convergence rate. The suggested method is also proven to identify unique sets of informative, relevant genes successfully with the highest overall average accuracy in 15 tested biological datasets. Additionally, the biological interpretations of the selected genes by the proposed method are also provided in our research work.

CHALLENGES OF DIGITAL ECONOMY AMONG HOUSEHOLDS IN UMUIKAA OKWU OLOKORO IN UMUAHIA SOUTH LOCAL GOVERNMENT AREA ABIA STATE

This study determined the extent to which households in Umuikaa participated in digital economy and illuminated on the challenges they encountered. The area of the study was Umuikaa Okwu Olokoro in Umuahia South Local Government Area Abia State Nigeria. The study adopted a descriptive research design. The population of the study was 97 households in Umuikaa. The study adopted a total population sampling. Two research objectives, two research questions and one hypothesis guided the study. The in.strument for data collection was a 20-item structured four points rating scale questionnaire. Three experts validated the instrument. A reliability of 0.86 was obtained using Cronbach Alpha Reliability Test Method. Findings of the study revealed that rural household in Umuikaa participated averagely in digital economy. The findings further revealed the challenges limiting rural household participation in digital economy. Among others were, cost of training on digitalization, lack of digital marketing skills, cost of digital tools and constantly changing technologies. The hypothesis revealed that 62.5% of the hotiseholds participated in digital economy. The study recommended that data which is the oil that drive digital economy should be subsidized by the government in order to make it affordable for rural households. Again, good networking without hitches should be installed in rural communities to encourage digital economy participation and rural inclusion among others.

A lightweight and personalized edge federated learning model

As a new distributed machine learning paradigm, federated learning has gained increasing attention in the industry and research community. However, federated learning is challenging to implement on edge devices with limited resources and heterogeneous data. This study aims to realize a lightweight and personalized model through pruning and masking with insufficient resources and heterogeneous data. Particularly, the server first downloads the subnetwork to the client according to the mask, and client prunes the subnetwork with the alternating direction method of multipliers (ADMM), so as to remove the unimportant parameters and reduce the cost of training and communication. At the same time, mask is used to mark the pruning condition of the model. Then, the unpruned parts and masks of local models are transmitted to the server for aggregation. The experimental results showed that the accuracy of the proposed model was improved by 9.36%, and the communication cost was reduced by 1.45 times compared with state-of-the-art models. Last but not least, we deploy flower identification models in Android Studio to illustrate the practicality of the proposed method.

Hebbian dreaming for small datasets

The dreaming Hopfield model constitutes a generalization of the Hebbian paradigm for neural networks, that is able to perform on-line learning when “awake” and also to account for off-line “sleeping” mechanisms. The latter have been shown to enhance storing in such a way that, in the long sleep-time limit, this model can reach the maximal storage capacity achievable by networks equipped with symmetric pairwise interactions. In this paper, we inspect the minimal amount of information that must be supplied to such a network to guarantee a successful generalization, and we test it both on random synthetic and on standard structured datasets (i.e., MNIST, Fashion-MNIST and Olivetti). By comparing these minimal thresholds of information with those required by the standard (i.e., always “awake”) Hopfield model, we prove that the present network can save up to ∼90% of the dataset size, yet preserving the same performance of the standard counterpart. This suggests that sleep may play a pivotal role in explaining the gap between the large volumes of data required to train artificial neural networks and the relatively small volumes needed by their biological counterparts. Further, we prove that the model Cost function (typically used in statistical mechanics) admits a representation in terms of a standard Loss function (typically used in machine learning) and this allows us to analyze its emergent computational skills both theoretically and computationally: a quantitative picture of its capabilities as a function of its control parameters is achieved and consistency between the two approaches is highlighted.The resulting network is an associative memory for pattern recognition tasks that learns from examples on-line, generalizes correctly (in suitable regions of its control parameters) and optimizes its storage capacity by off-line sleeping: such a reduction of the training cost can be inspiring toward sustainable AI and in situations where data are relatively sparse.

Finger Vein Identification Based on Large Kernel Convolution and Attention Mechanism.

FV (finger vein) identification is a biometric identification technology that extracts the features of FV images for identity authentication. To address the limitations of CNN-based FV identification, particularly the challenge of small receptive fields and difficulty in capturing long-range dependencies, an FV identification method named Let-Net (large kernel and attention mechanism network) was introduced, which combines local and global information. Firstly, Let-Net employs large kernels to capture a broader spectrum of spatial contextual information, utilizing deep convolution in conjunction with residual connections to curtail the volume of model parameters. Subsequently, an integrated attention mechanism is applied to augment information flow within the channel and spatial dimensions, effectively modeling global information for the extraction of crucial FV features. The experimental results on nine public datasets show that Let-Net has excellent identification performance, and the EER and accuracy rate on the FV_USM dataset can reach 0.04% and 99.77%. The parameter number and FLOPs of Let-Net are only 0.89M and 0.25G, which means that the time cost of training and reasoning of the model is low, and it is easier to deploy and integrate into various applications.

A Node Selection Strategy in Space-Air-Ground Information Networks: A Double Deep Q-Network Based on the Federated Learning Training Method

The Space-Air-Ground Information Network (SAGIN) provides extensive coverage, enabling global connectivity across a diverse array of sensors, devices, and objects. These devices generate large amounts of data that require advanced analytics and decision making using artificial intelligence techniques. However, traditional deep learning approaches encounter drawbacks, primarily, the requirement to transmit substantial volumes of raw data to central servers, which raises concerns about user privacy breaches during transmission. Federated learning (FL) has emerged as a viable solution to these challenges, addressing both data volume and privacy issues effectively. Nonetheless, the deployment of FL faces its own set of obstacles, notably the excessive delay and energy consumption caused by the vast number of devices and fluctuating channel conditions. In this paper, by considering the heterogeneity of devices and the instability of the network state, the delay and energy consumption models of each round of federated training are established. Subsequently, we introduce a strategic node selection approach aimed at minimizing training costs. Building upon this, we propose an innovative, empirically driven Double Deep Q Network (DDQN)-based algorithm called low-cost node selection in federated learning (LCNSFL). The LCNSFL algorithm can assist edge servers in selecting the optimal set of devices to participate in federated training before the start of each round, based on the collected system state information. This paper culminates with a simulation-based comparison, showcasing the superior performance of LCNSFL against existing algorithms, thus underscoring its efficacy in practical applications.

How Implementing a Digital Competency Management System Reduced Nurse Training Cost and Improved NPD Practitioner Satisfaction in a Pediatric Hospital.

Traditional competency evaluations and manual tracking created a problem within one organization. Subjective competency assessments and inconsistent recordkeeping resulted in increased organizational costs. These factors increased the workload of nursing professional development (NPD) practitioners, leading to job dissatisfaction and turnover. The project lead evaluated how implementing a digital competency management system (CMS) affected nurse training costs and assessed NPD practitioners' satisfaction after the digital CMS conversion. A cost analysis compared training costs before and after implementation of a digital CMS. A pre- and postsurvey compared NPD practitioners' satisfaction before and after digital implementation. A digital CMS provided a centralized repository and educational platform, which saved $500,000 in training costs and increased NPD practitioners' job satisfaction by decreasing workload and creating meaningful mechanisms to accurately assess staff knowledge and skills. The NPD practitioners showed tangible outcome measures and return on investment by combining adult learning and technology. [J Contin Educ Nurs. 2024;55(5):239-245.].

Lightweight Imitation Learning for Real-Time Cooperative Service Migration

Due to the revolution of communication technology, the rapidly increasing number of mobile devices in edge networks generates various real-time service requests, requiring a considerable volume of heterogeneous resources all the time. However, edge devices with limited resources cannot afford substantial learning cost, while migrating services requires heterogeneous resources, especially for dynamic networks. To address these issues, we first establish a cooperative service migration framework and formulate a bi-objective optimization problem to optimize service performance and cost. By analyzing the optimal migration ratio of service cooperative migration, we propose an offline expert policy based on global states to provide optimal expert demonstrations. To realize real-time service migration based on observable states, we design a lightweight online agent policy to imitate expert demonstrations and leverage meta update to accelerate the model transfer. Experimental results show that our algorithm is exceptional in training cost and accuracy, and has significant superiors in multiple metrics such as the service latency and payment under different workloads, compared to other representative algorithms.

The impact of robotic rectal cancer surgery at a Canadian regional cancer centre: a retrospective cohort study.

Although robotic surgery has several advantages over other minimally invasive surgery (MIS) techniques for rectal cancer surgery, the uptake in Canada has been limited owing to a perceived increase in cost and lack of training. The objective of this study was to determine the impact of access to robotic surgery in a Canadian setting. We conducted a retrospective cohort study involving consecutive adults undergoing surgical resection for rectal cancer between 2017 and 2020. The primary exposure was access to robotic surgery. Outcomes included MIS utilization, short-term outcomes, total cost of care, and quality of surgical resection. We completed univariate and multivariate analyses. We included 171 individuals in this cohort study (85 in the prerobotic period and 86 in the robotic period). The 2 groups had similar baseline characteristics. A higher proportion of individuals underwent successful MIS in the robotic phase (86% v. 46%, p < 0.001). Other benefits included a shorter mean length of hospital stay (5.1 d v. 9.2 d, p < 0.001). The quality of surgical resection was similar between groups. The total cost of care was $16 746 in the robotic period and $18 808 in the prerobotic period (mean difference -$1262, 95% confidence interval -$4308 to $1783; p = 0.4). Access to robotic rectal cancer surgery increased successful completion of MIS and shortened hospital stay, with a similar total cost of care. Robotic rectal cancer surgery can enhance patient outcomes in the Canadian setting.

How Can We Train Deep Learning Models Across Clouds and Continents? An Experimental Study

This paper aims to answer the question: Can deep learning models be cost-efficiently trained on a global market of spot VMs spanning different data centers and cloud providers? To provide guidance, we extensively evaluate the cost and throughput implications of training in different zones, continents, and clouds for representative CV, NLP and ASR models. To expand the current training options further, we compare the scalability potential for hybrid-cloud scenarios by adding cloud resources to on-premise hardware to improve training throughput. Finally, we show how leveraging spot instance pricing enables a new cost-efficient way to train models with multiple cheap VMs, trumping both more centralized and powerful hardware and even on-demand cloud offerings at competitive prices.

DAHA: Accelerating GNN Training with Data and Hardware Aware Execution Planning

Graph neural networks (GNNs) have been gaining a reputation for effective modeling of graph data. Yet, it is challenging to train GNNs efficiently. Many frameworks have been proposed but most of them suffer from high batch preparation cost and data transfer cost for mini-batch training. In addition, existing works have limitations on the device utilization pattern, which results in fewer opportunities for pipeline parallelism. In this paper, we present DAHA, a GNN training framework with data and hardware aware execution planning to accelerate end-to-end GNN training. We first propose a data and hardware aware cost model that is lightweight and gives accurate estimates on per-operation time cost for arbitrary input and hardware settings. Based on the cost model, we further explore the optimal execution plan for the data and hardware with three optimization strategies with pipeline parallelism: (1) group-based in-turn pipelining of batch preparation neural training to explore more optimization opportunities and prevent batch preparation bottlenecks; (2) data and hardware aware rewriting for intra-batch execution planning to improve computation efficiency and create more opportunities for pipeline parallelism; and (3) inter-batch scheduling to further boost the training efficiency. Extensive experiments demonstrate that DAHA can consistently and significantly accelerate end-to-end GNN training and generalize to different message-passing GNN models.

Standardized patients’ experiences of portraying characters in difficult communication scenarios: Narrative inquiry

Background There are limited studies that explored the preparation and challenges faced by standardized patients (SPs) in portraying characters in difficult communication scenarios, and the strategies used to overcome these challenges. The purpose of this study was to understand the experience of SPs in interpreting difficult communication situations and the learning needs of performing similar scenarios. And it allows the researchers to explore the meaning, beliefs, values, and aspiration associated with their role as SPs. The findings could shade light on the significance of their experiences and provide valuable insights for the development of future SP training programs. Methods The design of this study is framed by a narrative inquiry, using semi-structured guidelines to conduct in-depth interviews with 11 SPs who have participated in the performances of difficult communication situations. Research data were analyzed by Polkinghorne narrative analysis, and Riessman’s four criteria were used to establish rigor. Results Analysis revealed the following five themes: scenarios to real life connections, process of preparing for a performance, methods to detach from character, obtaining unexpected rewards, and needs for performance training. There are two to three subthemes that are subsumed under each theme. Conclusions To strengthen training in difficult communication for healthcare professionals, the use of SPs to interpret challenging difficult communication scenarios will continue to increase. Educators need to ensure that SPs are fully prepared physically and emotionally before, during and after their performance. Offering of continuing education and training in feedback techniques are crucial to extend the tenure of SPs, reduce their frustration, prevent attrition, and ultimately, reduce training costs. In the future, SP training should also include detachment and feedback techniques to alleviate SPs’ stress.

Toward grouped-reservoir computing: organic neuromorphic vertical transistor with distributed reservoir states for efficient recognition and prediction

Reservoir computing has attracted considerable attention due to its low training cost. However, existing neuromorphic hardware, focusing mainly on shallow-reservoir computing, faces challenges in providing adequate spatial and temporal scales characteristic for effective computing. Here, we report an ultra-short channel organic neuromorphic vertical transistor with distributed reservoir states. The carrier dynamics used to map signals are enriched by coupled multivariate physics mechanisms, while the vertical architecture employed greatly increases the feedback intensity of the device. Consequently, the device as a reservoir, effectively mapping sequential signals into distributed reservoir state space with 1152 reservoir states, and the range ratio of temporal and spatial characteristics can simultaneously reach 2640 and 650, respectively. The grouped-reservoir computing based on the device can simultaneously adapt to different spatiotemporal task, achieving recognition accuracy over 94% and prediction correlation over 95%. This work proposes a new strategy for developing high-performance reservoir computing networks.

Separable ConvNet Spatiotemporal Mixer for Action Recognition

Video action recognition is vital in the research area of computer vision. In this paper, we develop a novel model, named Separable ConvNet Spatiotemporal Mixer (SCSM). Our goal is to develop an efficient and lightweight action recognition backbone that can be applied to multi-task models to increase the accuracy and processing speed. The SCSM model uses a new hierarchical spatial compression, employing the spatiotemporal fusion method, consisting of a spatial domain and a temporal domain. The SCSM model maintains the independence of each frame in the spatial domain for feature extraction and fuses the spatiotemporal features in the temporal domain. The architecture can be adapted to different frame rate requirements due to its high scalability. It is suitable to serve as a backbone for multi-task video feature extraction or industrial applications with its low prediction and training costs. According to the experimental results, SCSM has a low number of parameters and low computational complexity, making it highly scalable with strong transfer learning capabilities. The model achieves video action recognition accuracy comparable to state-of-the-art models with a smaller parameter size and fewer computational requirements.

One-shot sim-to-real transfer policy for robotic assembly via reinforcement learning with visual demonstration

AbstractReinforcement learning (RL) has been successfully applied to a wealth of robot manipulation tasks and continuous control problems. However, it is still limited to industrial applications and suffers from three major challenges: sample inefficiency, real data collection, and the gap between simulator and reality. In this paper, we focus on the practical application of RL for robot assembly in the real world. We apply enlightenment learning to improve the proximal policy optimization, an on-policy model-free actor-critic reinforcement learning algorithm, to train an agent in Cartesian space using the proprioceptive information. We introduce enlightenment learning incorporated via pretraining, which is beneficial to reduce the cost of policy training and improve the effectiveness of the policy. A human-like assembly trajectory is generated through a two-step method with segmenting objects by locations and iterative closest point for pretraining. We also design a sim-to-real controller to correct the error while transferring to reality. We set up the environment in the MuJoCo simulator and demonstrated the proposed method on the recently established The National Institute of Standards and Technology (NIST) gear assembly benchmark. The paper introduces a unique framework that enables a robot to learn assembly tasks efficiently using limited real-world samples by leveraging simulations and visual demonstrations. The comparative experiment results indicate that our approach surpasses other baseline methods in terms of training speed, success rate, and efficiency.