Performance Of Specific Tasks Research Articles

Automatic movie genre classification & emotion recognition via a BiProjection Multimodal Transformer

Analyzing, manipulating, and comprehending data from multiple sources (e.g., websites, software applications, files, or databases) and of diverse modalities (e.g., video, images, audio and text) has become increasingly important in many domains. Despite recent advances in multimodal classification (MC), there are still several challenges to be addressed, such as: the combination of modalities of very diverse nature, the optimal feature engineering for each modality, as well as the semantic alignment between text and images. Accordingly, the main motivation of our research relies in devising a neural architecture that effectively processes and combines text, image, video and audio modalities, so it can offer a noteworthy performance in different MC tasks. In this regard, the Multimodal Transformer (MulT) model is a cutting-edge approach often employed in multimodal supervised tasks, which, although effective, has the problem of having a fixed architecture that limits its performance in specific tasks as well as its contextual understanding, meaning it may struggle to capture fine-grained temporal patterns in audio or effectively model spatial relationships in images. To address these issues, our research modifies and extends the MulT model in several aspects. Firstly, we focus on leveraging the Gated Multimodal Unit (GMU) module within the architecture to efficiently and dynamically weigh modalities at the instance level and to visualize the use of modalities. Secondly, to overcome the problem of vanishing and exploding gradients we focus on strategically placing residual connections in the architecture. The proposed architecture is evaluated in two different and complex classification tasks, on the one hand, the movie genre categorization (MGC) and, on the other hand, the multimodal emotion recognition (MER). The results obtained are encouraging as they indicate that the proposed architecture is competitive against state-of-the-art (SOTA) models in MGC, outperforming them by up to 2% on the Moviescope dataset, and by 1% on the MM-IMDB datasets. Furthermore, in the MER task the unaligned version of the datasets was employed, which is considerably more difficult; we improve accuracy SOTA results by up to 1% on the IEMOCAP dataset, and attained a competitive outcome on the CMU-MOSEI11Dai et al., (2021) collection, outperforming SOTA results in several emotions.

Advancing language models through domain knowledge integration: a comprehensive approach to training, evaluation, and optimization of social scientific neural word embeddings

This article proposes a comprehensive strategy for training, evaluating, and optimizing domain-specific word2vec-based word embeddings, using social science literature as an example. Our primary objectives are: (1) to train the embeddings utilizing a corpus of social science text, (2) to test their performance against domain-unspecific embeddings using our developed intrinsic and extrinsic evaluation strategy, and (3) to enhance their performance even further by using domain knowledge. As an integral part of this approach, we present SociRel-461, a domain-knowledge dictionary designed for the intrinsic evaluation and subsequent refinement of social science word embeddings. Using a dataset of 100,000 full-text scientific articles in sociology, we train multiple vector space models, which we then benchmark against a larger, pre-trained general language embedding model as part of our extrinsic evaluation. Furthermore, we developed a transfer learning multi-label classification task for extrinsic evaluation. Our findings reveal that domain-specific embeddings outperform their domain-unspecific counterparts in both intrinsic and extrinsic evaluations. We also investigated the retrofitting post-processing method to enhance domain-unspecific embeddings with the domain knowledge embedded in SociRel-461. While retrofitting does not enhance our domain-specific vector space models, it significantly improves the performance of the domain-unspecific embeddings. This highlights the potential of retrofitting for the transfer of domain knowledge to domain-unspecific embeddings. Our results emphasize the importance of utilizing domain-specific word embeddings for better performance in domain specific transfer learning tasks, as they outperform conventional embeddings trained on everyday language.

HGR Correlation Pooling Fusion Framework for Recognition and Classification in Multimodal Remote Sensing Data

This paper investigates remote sensing data recognition and classification with multimodal data fusion. Aiming at the problems of low recognition and classification accuracy and the difficulty in integrating multimodal features in existing methods, a multimodal remote sensing data recognition and classification model based on a heatmap and Hirschfeld–Gebelein–Rényi (HGR) correlation pooling fusion operation is proposed. A novel HGR correlation pooling fusion algorithm is developed by combining a feature fusion method and an HGR maximum correlation algorithm. This method enables the restoration of the original signal without changing the value of transmitted information by performing reverse operations on the sample data. This enhances feature learning for images and improves performance in specific tasks of interpretation by efficiently using multi-modal information with varying degrees of relevance. Ship recognition experiments conducted on the QXS-SROPT dataset demonstrate that the proposed method surpasses existing remote sensing data recognition methods. Furthermore, land cover classification experiments conducted on the Houston 2013 and MUUFL datasets confirm the generalizability of the proposed method. The experimental results fully validate the effectiveness and significant superiority of the proposed method in the recognition and classification of multimodal remote sensing data.

Evaluating the language abilities of Large Language Models vs. humans: Three caveats

We identify and analyze three caveats that may arise when analyzing the linguistic abilities of Large Language Models. The problem of unlicensed generalizations refers to the danger of interpreting performance in one task as predictive of the models’ overall capabilities, based on the assumption that because a specific task performance is indicative of certain underlying capabilities in humans, the same association holds for models. The human-like paradox refers to the problem of lacking human comparisons, while at the same time attributing human-like abilities to the models. Last, the problem of double standards refers to the use of tasks and methodologies that either cannot be applied to humans or they are evaluated differently in models vs. humans. While we recognize the impressive linguistic abilities of LLMs, we conclude that specific claims about the models’ human-likeness in the grammatical domain are premature.

Pediatric Performance-Based Outcome Measures for Upper Extremity Function: A Scoping Review and Linking to the International Classification of Functioning, Disability, and Health.

Background. Performance-based outcome measures (PBOMs) are objective measures that assess physical capacity or performance in specific tasks or movements. Purpose. 1) to identify which PBOMs are most frequently reported to evaluate upper extremity (UE) function in pediatric rehabilitation 2) to determine the link between constructs of the ICF and meaningful concepts extracted from each identified PBOM. Methods. Pediatric UE PBOMs were searched in four databases. The selection of outcome measures included an initial title and abstract screening, followed by full-text review of the articles to be included based on identified selection criteria. Two reviewers were appointed to link the meaningful concepts identified in the outcome measures independently and a third reviewer was consulted in case of ambiguity to make a final decision. Findings. After the initial screening, 1786 full-text articles were reviewed, 1191 met the inclusion criteria, in which 77 outcome measures were identified and 32 were included in the linking process. From the included 32 outcome measures, 538 items were extracted and linked to the ICF. The most commonly cited measures included Assisting Hand Assessment, Jebsen-Taylor Hand Function Test, Melbourne Assessment of Unilateral Upper Limb. The Activity and Participation domain represented 364 codes followed by the Body Functions domain domain which represented 174 codes. Implications. A majority of the outcome measures identified were linked with the Mobility, Fine Hand Use of the ICF. Therefore, when selecting a PBOM, careful considerations need to be made regarding which concept of health is to be assessed.

Research on Improved Algorithm for Small Object Detection in Intelligent Surveillance Video based on YOLOv7

In order to address the issue of small objects being difficult to detect effectively in intelligent surveillance videos, this study proposes an improved scheme for the YOLOv7-tiny algorithm. This scheme integrates the Convolutional Block Attention Module (CBAM) into YOLOv7-tiny, effectively enhancing the model's feature extraction and small object detection capabilities in complex backgrounds, thereby improving the overall detection precision. Experimental evaluations indicate that the improved algorithm shows enhanced performance in specific small object detection tasks, achieving an accuracy of 85.6%, a recall rate of 85.2%, and a mean average precision (mAP) of 90.2%. These results demonstrate the effectiveness and practical value of the improved scheme in enhancing the performance of YOLOv7-tiny in small object detection tasks.

Application of transcranial alternating current stimulation to improve eSports-related cognitive performance.

Electronic Sports (eSports) is a popular and still emerging sport. Multiplayer Online Battle Arena (MOBA) and First/Third Person Shooting Games (FPS/TPS) require excellent visual attention abilities. Visual attention involves specific frontal and parietal areas, and is associated with alpha coherence. Transcranial alternating current stimulation (tACS) is a principally suitable tool to improve cognitive functions by modulation of regional oscillatory cortical networks that alters regional and larger network connectivity. In this single-blinded crossover study, 27 healthy college students were recruited and exposed to 10 Hz tACS of the right frontoparietal network. Subjects conducted a Visual Spatial Attention Distraction task in three phases: T0 (pre-stimulation), T1 (during stimulation), T2 (after-stimulation), and an eSports performance task which contained three games ("Exact Aiming," "Flick Aiming," "Press Reaction") before and after stimulation. The results showed performance improvements in the "Exact Aiming" task and hint for a prevention of reaction time performance decline in the "Press Reaction" task in the real, as compared to the sham stimulation group. We also found a significant decrease of reaction time in the visual spatial attention distraction task at T1 compared to T0 in the real, but not sham intervention group. However, accuracy and inverse efficiency scores (IES) did not differ between intervention groups in this task. These results suggest that 10 Hz tACS over the right frontal and parietal cortex might improve eSports-related skill performance in specific tasks, and also improve visual attention in healthy students during stimulation. This tACS protocol is a potential tool to modulate neurocognitive performance involving tracking targets, and might be a foundation for the development of a new concept to enhance eSports performance. This will require however proof in real life scenarios, as well optimization.

DCCN: A dual-cross contrastive neural network for 3D point cloud representation learning

The proliferation of depth cameras and LiDAR sensors in actual industrial environments has fueled the pursuit of an effective and efficient 3D point cloud model that enables us to perceive and interact with the physical world. However, the intrinsic complexity of 3D semantic information poses significant challenges to model design, including spatial rotation invariance and irregular point cloud structure, which fundamentally impact the representation and behavior of 3D point cloud systems. Existing have either heavily relied on labeling information in a supervised learning setting or failed to effectively capture the inherent patterns of the 3D point clouds within a self-supervised learning framework, leading to poor performance in specific downstream tasks. To address these limitations, this paper introduces a self-supervised framework, Dual-Cross Contrastive Neural Network (DCCN) for 3D point cloud self-supervised representation learning. DCCN leverages cross-view, cross-network, and domain-specific knowledge distillation to enhance the extraction of hidden features from point clouds and fully exploit the capabilities of the encoder. Our DCCN employs a pseudo-Siamese network consisting of an online network and a target network, facilitating knowledge interaction and distillation. The method extracts internal states from augmented 3D point cloud by learning cross-view relationships and optimizes model parameters through intra-modal cross-network learning. We incorporate a momentum-updating mechanism without shared weights in the Siamese network architecture to distill knowledge and enhance the role differentiation the online and target networks. Experimental results demonstrate that our approach outperforms a range of supervised and self-supervised learning methods across a series of downstream tasks consisting of four tasks in three representative datasets. Ablation studies validate the component-wise effectiveness of cross-view, cross-network, and moment-updating learning objectives in achieving superior point cloud representation. The overall findings establish our method, DCCN, as an effective solution for 3D point cloud representation learning in real-world applications.

Generative pretrained transformer-4, an artificial intelligence text predictive model, has a high capability for passing novel written radiology exam questions.

AI-image interpretation, through convolutional neural networks, shows increasing capability within radiology. These models have achieved impressive performance in specific tasks within controlled settings, but possess inherent limitations, such as the inability to consider clinical context. We assess the ability of large language models (LLMs) within the context of radiology specialty exams to determine whether they can evaluate relevant clinical information. A database of questions was created with official sample, author written, and textbook questions based on the Royal College of Radiology (United Kingdom) FRCR 2A and American Board of Radiology (ABR) Certifying examinations. The questions were input into the Generative Pretrained Transformer (GPT) versions 3 and 4, with prompting to answer the questions. One thousand seventy-two questions were evaluated by GPT-3 and GPT-4. 495 (46.2%) were for the FRCR 2A and 577 (53.8%) were for the ABR exam. There were 890 single best answers (SBA), and 182 true/false questions. GPT-4 was correct in 629/890 (70.7%) SBA and 151/182 (83.0%) true/false questions. There was no degradation on author written questions. GPT-4 performed significantly better than GPT-3 which selected the correct answer in 282/890 (31.7%) SBA and 111/182 (61.0%) true/false questions. Performance of GPT-4 was similar across both examinations for all categories of question. The newest generation of LLMs, GPT-4, demonstrates high capability in answering radiology exam questions. It shows marked improvement from GPT-3, suggesting further improvements in accuracy are possible. Further research is needed to explore the clinical applicability of these AI models in real-world settings.

A comprehensive investigation for ChatGPTs applications in education

With the surge in popularity of Chat Generative Pre-trained Transformer (ChatGPT), an artificial intelligence technology, on a global scale, numerous industries have started to take notice of its potential applications in their specific sectors, including the field of education. This document presents a thorough examination of how ChatGPT is being used in the realm of education. The following piece outlines the approaches to integrating ChatGPT into education, viewed from three angles: Educational Assessment, Exercise Generation, and Personalized Learning. It also provides comprehensive guidelines aimed at both teachers and students. Utilizing ChatGPT in education allows for personalized one-on-one assistance to students, addressing challenges posed by uneven distribution of educational resources and individual differences among students. Current research on applying ChatGPT in education mostly revolves around designing prompts to elicit specific responses or using generated content for evaluating its subject-specific capabilities. Given that ChatGPT is a more versatile form of artificial intelligence not limited to any particular domain, the article suggests that future research should focus more on integrating ChatGPT into existing educational systems, ensuring its consistent performance in specific tasks. Furthermore, considering ChatGPT's possibility of generating imprecise information, it is necessary to explore how to fine-tune ChatGPT using specialized databases, enabling it to generate reliable content in professional subject areas.

Exploring the Implications of Inter-Limb Asymmetries on Sprint, Agility, and Jump Performance in Young Highly-Trained Basketball Athletes: Is There a Relevant Threshold?

Background and Objectives: This study aimed to investigate the magnitude of vertical jump inter-limb asymmetries among young highly-trained basketball athletes and to analyze its impact on sport performance, specifically in sprints, agility, and vertical jumps. Materials and Methods: A unilateral countermovement jump (CMJ) was employed to determine Inter-limb Index Asymmetry (IAI) in 320 participants aged from 14 to 18 years, from the Valencia Basket youth academy. IAI was categorized into three groups: 0-9.9%, 10-14.9%, and >15%. The relationship between IAI and performance variables was analyzed through correlation studies (Pearson or Spearman's). The influence of IAI magnitude was assessed using ANOVA or Kruskal-Wallis analysis, with leg dominance as a covariable. SPSS Statistics version 26 was used for analysis. Results: Among all the participants, the mean IAI was 10.6%. Correlation studies revealed non-significant values (p < 0.05) between IAI and sport performance variables. The three IAI magnitude groups did not show statistically significant differences in sprint, agility, and jump outcomes. Leg dominance did not seem to influence performance outcomes, except for unilateral CMJ. Conclusions: The results obtained challenge the assumption that an IAI above 10% negatively affects sprint, agility, or jump performance in young basketball athletes. Notably, the magnitude of IAI did not influence sport performance parameters, suggesting that the 10-15% threshold from previous studies may not be applicable to this population. The study emphasizes the need to understand lower-limb asymmetries in the context of specific sport task performance, considering the potential evolution over time among affected young athletes.

Evaluation of the relationship between occupational-specific task performance and measures of physical fitness, cardiovascular and musculoskeletal health in firefighters

IntroductionFirefighters are required to perform physically strenuous tasks such as hose drags, victim rescues, forcible entries and stair climbs to complete their public safety mission. Occupational-specific tasks are often used to evaluate the ability of firefighters to adequately/safely perform their duties. Depending on the regions, occupational-specific tasks include six to eight individual tasks, which emphasize distinct aspects of their physical fitness, while also requiring different levels of cardiovascular (CVH) and musculoskeletal health (MSH). Therefore, the aim of this study was to evaluate the relationship between specific occupational task performance and measures of physical fitness, cardiovascular and musculoskeletal health.MethodsUsing a cross-sectional design, 282 full-time male and female firefighters were recruited. A researcher-generated questionnaire and physical measures were used to collect data on sociodemographic characteristics, CVH, MSH and weekly physical activity habits. Physical measures were used to collect data on physical fitness and occupational-specific task performance.ResultsAbsolute cardiorespiratory fitness (abV̇O2max), grip strength, leg strength, push-ups, sit-ups and lean body mass (all p < 0.001) had an inverse association with completion times on all occupational-specific tasks. Age was positively related to the performance of all tasks (all p < 0.05). Higher heart rate variability (HRV) was associated with better performance on all tasks (all p < 0.05). Bodyfat percentage (BF%) and diastolic blood pressure were positively associated with the step-up task (p < 0.05). Lower back musculoskeletal injury (LoBMSI), musculoskeletal discomfort (MSD), and lower limb MSD were associated with a decreased odds of passing the step-up. Upper body MSIs (UBMSI), LoBMSIs and Lower back MSD were associated with decreased odds of passing the rescue drag.ConclusionFirefighters that were taller, leaner, stronger and fitter with a more favourable CVH profile, higher HRV and less musculoskeletal discomfort performed best on all occupational-specific tasks.

Leaders' Machiavellianism and subordinates' task performance: a multilevel investigation

PurposeThis study aims at analyzing the detrimental cross-level serial effects of leaders' Machiavellianism and abusive supervision on subordinates' burnout and task performance. The general aggression model and conservation of resources theory guide our research model.Design/methodology/approachThe authors developed a multilevel design and used multisource data. The authors collected data from 50 bank branches located in Pakistan. A total of 50 branch supervisors participated, which yielded 200 supervisor-subordinate dyads. Machiavellianism was self-assessed by the supervisors who further rated the specific task performance of each of their respective subordinates. Burnout and abusive supervision ratings were provided by the subordinates. Abusive supervision scores were aggregated at the group level.FindingsMachiavellianism has an indirect negative effect on individual-level task performance through the serial cross-level mediating effects of abusive supervision at the group level and attendant individual-level burnout.Practical implicationsFindings provide practical implications for the management of task performance and human resources.Originality/valueThis study offers a comprehensive cross-level model to analyze the effect of Machiavellianism on group-level and individual-level outcomes known to affect the effectiveness of leaders.

Somatotypes of Top Croatian Male Volleyball Players

Situational efficiency of volleyball players depends on many factors, including specific ones in the area of morphology. The specialization of playing roles implies the performance of specific motor tasks, and therefore a specific combination of morphological factors necessary for the player’s efficiency. This study aimed to determine the somatotypes of groups of top-level volleyball players divided according to their playing role. The research was conducted on 40 (body height 192.22±8.58 cm, body weight 86.93±11.22 kg, age 24.18±4.97 years) male volleyball players from the Croatian A1 volleyball league. According to the playing role, 10 setters (body height 191.21±6.39 cm, body weight 84.12±8.76 kg, age 24.12±6.77 years), 10 middle blockers (body height 199.97±4.41 cm, body weight 94.82±6.67 kg, age 26.05±26.05 years), 10 liberos (body height 181.71±5.71 cm, body weight 74.97.21±5.59 kg, age 22.67±2.88 years) and 10 outside hitters (body height 196.06±4.34 cm, body weight 93.80±10.29 kg, age 23.89±5.53 years). Heath-Carter method was used for the somatotype determination. A statistically significant difference between the groups was detected in Body height, Body weight, Flexed arm girth circumference, Flexed calf girth circumference and Calf skinfold. The determined somatotypes for setters (2.8 - 3.8 - 3.4) and middle blockers (2.3 - 3.8 - 3.6) are mesomorph-ectomorph, and liberos (2.4 - 4.6 - 3.0) and outside hitters (2.5 - 4.3 - 3.1) are ecto-mesomorph. The results are in line with previous research. This confirms the importance of the longitudinal dimensionality of the skeleton in volleyball. In the future, we can expect the growth of the ectomorph component, with a stagnation or slight increase in the mesomorph component of the somatotype in top level volleyball.

Google Trends as a digital marketing tool: methodical aspects

The aim of the article is to identify and analyze the menus and options of Google Trends for the digital marketers. Analyses results. The article is devoted to the problem of expanding the information base of marketers through the use of Google Trends, which allows to analyze the search queries of users of the Google search system and create appropriate graphic visualizations of the popularity of various search terms and topics. It was established that the researched tool expands the information base for the work of digital marketers, in particular through the use of such sections of the service as «Explore», «Compare», «Trending Searchers. Daily search trends», «Trending Searchers. Realtime Search Trends», «Year in Search», «Subscriptions». For example, data from the «Year in search» category, in particular the «purchase of the year» category, shows the most relevant user needs that were relevant in a certain year. The general and additional capabilities of Google Trends for digital marketers have been highlighted. The general capabilities of this service include: determining trends by specific search terms or topics, determining the specifics of the dynamics of the popularity of specific search terms or topics, searching and analyzing competitors, researching various digital search platforms, and operational monitoring general current trends in the country. The general capabilities will be useful for all digital marketers, regardless of the specifics of their activity. Additional features of Google Trends include: creation of relevant content, refinement of the semantic core of the site, optimization of the content plan, clustering of topics and the search for reasons for website traffic fluctuations. Additional opportunities will be useful for more highly specialized product promotion specialists (content marketers, brand marketers, SEO specialists, SMM marketers, story makers, etc.). The authors summarized that the research of the dynamics of interest of network users through Google Trends helps to plan marketing initiatives for the future, align the content of sites with current trends, and analyze the activities of competitors. Conclusions and directions for further research. The analysis of the features of the use of other digital marketing tools and the comparison between them in terms of the effectiveness of their use for the performance of specific tasks set before digital marketers are identified as promising for further research.

ОРГАНІЗАЦІЯ ОХОРОНИ Й ОБОРОНИ БАЗОВОГО ТАБОРУ ВІЙСЬКОВИМИ ЧАСТИНАМИ (ПІДРОЗДІЛАМИ) НАЦІОНАЛЬНОЇ ГВАРДІЇ УКРАЇНИ ДЛЯ ВИКОНАННЯ ЗАВДАНЬ ІЗ ЗАБЕЗПЕЧЕННЯ ДЕРЖАВНОЇ БЕЗПЕКИ

During the years of Ukraine's independence, servicemen of the Armed Forces of Ukraine as part of military units (units) and peacekeeping contingents repeatedly participated in the performance of specific tasks in peacekeeping operations in the territories of the Republic of Iraq, the Democratic Republic of the Congo, Ivory Coast, Sierra Leone, and Liberia, Kosovo. Thus, a certain amount of experience was accumulated in the performance of various tasks in the territories of other states, issues of unit management, organization of interaction and cooperation with other contingents and local authorities were worked out, a wide range of tasks were solved, including the preparation and order of occupation of base camps, their arrangement, organizations of the protection and defense system. In the period from 2014, the issue of protection and defense of base camps in the area of the anti-terrorist operation, which changed from April 30, 2018 to the operation of the United Forces, and from February 24, 2022, turned into a full-scale aggression on the part of the Russian Federation. The procedure for organizing the day-to-day activities of military units (subunits) of the Armed Forces of Ukraine and the National Guard of Ukraine, which are located in base camps, has become very relevant. The analysis of recent events shows that commanders (supervisors) do not always pay enough attention to issues of protection and defense of base camps, organization of daily activities in them. Therefore, it is important to reveal the issues of preparation and the procedure for occupying base camps, the creation of an effective system of their protection and defense, and the organization of daily activities. In order to ensure favorable conditions for protection and defense against the ground and air enemy and the vital activities of the troops, organize and equip the main facilities of the base camp in terms of engineering (locations of personnel, weapons and military equipment, warehouses of material resources). The protection of the identified objects should be carried out by placing them in capital buildings, creating protective earthen ramparts, equipping shelters for personnel and equipment, using the natural properties of the area, etc.

ЗАДАЧА РОЗПОДІЛУ ПРЕДМЕТІВ РЕЧОВОГО МАЙНА ТА ПРОДОВОЛЬЧОГО ЗАБЕЗПЕЧЕННЯ В ЕЛЕМЕНТАХ КОМПЛЕКСУ БОЙОВОГО ЕКІПІРУВАННЯ ВІЙСЬКОВОСЛУЖБОВЦЯ НАЦІОНАЛЬНОЇ ГВАРДІЇ УКРАЇНИ

The development of the military sphere of activity is ongoing and comprehensive. One of the areas of such development is the equipment of military personnel. Nowadays, much attention is paid to combat equipment in the armies of all countries of the world. According to the Concept for the Creation of a Combat Equipment Complex for the Armed Forces of Ukraine, combat equipment is a set of items that constitute the equipment of a serviceman during combat missions. A combat equipment complex is generally understood to mean the combat equipment of servicemen created to increase the effectiveness of their personal actions in combat (hostilities), as well as as part of a unit in the squad, platoon, company, and battalion. A modern combat equipment complex includes a large number of elements that are interconnected, and their interaction generates properties that are not inherent in its individual elements. Each element of this kit has its own weight, dimensions, and operational features, which together create a load on the serviceman, forming the distribution of mechanical forces and moments on the human body. Consequently, there is a contradiction between the growing capabilities provided by each element of the combat equipment complex for a soldier in combat and the limited physical capabilities of the soldier as a carrier of the equipment and its user. One of the ways to resolve this contradiction is to divide the elements of the combat equipment complex into echelons. Taking into account the differences in the content of tasks, duration of actions and conditions of their performance, it is important to create preconditions for achieving high efficiency of servicemen in the performance of each type of task. An important role in creating such prerequisites is played by the distribution of a set of items provided for by the supply standards by echelons. A rational distribution will allow a serviceman to carry those items that are necessary for life support during the performance of specific tasks, without overloading him with unnecessary (surplus) property. The task of echeloning the combat equipment complex of a serviceman of the NGU by elements of material and food supply can be formulated using a theoretical-multiple approach. In order to find effective solutions for the echeloning of items of clothing and food supply, it is necessary to identify and study the factors that affect the needs of a serviceman in combat equipment items for specific special (combat) tasks, to introduce appropriate indicators, selection criteria and quantitative restrictions imposed on the values of the echelons' parameters.

Wątpliwości i ryzyka wynikające ze współpracy gmin w ramach instytucji porozumienia międzygminnego

The article discusses the commencement of cooperation by communes in the form of intercommunal agreements, which in principle allow to optimize the implementation of statutory obligations related to the performance of specific public tasks by communes. Due to the rudimentary nature of regulations, both the doctrine and jurisprudence, there are many doubts related to the qualification of these agreements as institutions of civil or administrative law, including whether they are acts of local law. This results in risk on many levels, ranging from their correct announcement, through the pursuit of claims, to the end of their legal existence. This institution is a promising form that allows municipalities to perform their tasks based on the best effect mechanism by establishing cooperation between municipalities achieving results in the implementation of specific public tasks. In practice, it is not particularly popular due to a number of doubts and risks arising from the residual statutory regulations.

Методика формування артикуляційної моторики у дітей зі стертою формою дизартрії (індивідуально-зорієнтований підхід)

In the current conditions of introducing inclusion into the education system, individually oriented education is the leading direction that determines its further development. In this context, the formation of individual programs that ensure the development of children's speech is one of the priority areas of research. The article, based on the analysis of scientific and theoretical sources, defines the differentiated differences between the concepts of "dysarthria" and "erased form of dysarthria," characterizes the main approaches to the study of this speech disorder, classification, diagnostic orientation in its study, and directions of speech therapy work. It was determined that the nature and mechanisms of speech disorders in dysarthria are complex, multi-level, and multifactorial, which implies the need to use an individually oriented approach to overcoming this speech disorder. The method of distinguishing the types of articulatory motility disorders in dysarthria and the model of forming individual development programs for children with an erased form of articulatory motility dysarthria are presented. The research is based on individual, neurologopedic, and activity approaches. It was determined that applying these approaches allows for effectively rationalizing the speech therapy process to narrow it down to the performance of specific tasks. Under such conditions, the correction process is established from light to severe. Moreover, the training is generally carried out in a situation of success for the child because even the most difficult tasks, inaccessible to him at the beginning of the speech therapy work, he performs with ease, provided they are applied promptly, with little help. Speech therapy organized in this way brings satisfaction to the child. As a result of the conducted practical training, we concluded that each child with an erased form of dysarthria has its own pace of work. The differentiated symptomatology of the disorder is due to various reasons. In the process of speech therapy work, a correction path is built for each of the children, and different exercises and tasks are selected. In one speech therapy session, a variety of tasks can be used, clearly oriented towards the achievement of intermediate goals and the final goal.

Brain-inspired Predictive Coding Improves the Performance of Machine Challenging Tasks.

Backpropagation has been regarded as the most favorable algorithm for training artificial neural networks. However, it has been criticized for its biological implausibility because its learning mechanism contradicts the human brain. Although backpropagation has achieved super-human performance in various machine learning applications, it often shows limited performance in specific tasks. We collectively referred to such tasks as machine-challenging tasks (MCTs) and aimed to investigate methods to enhance machine learning for MCTs. Specifically, we start with a natural question: Can a learning mechanism that mimics the human brain lead to the improvement of MCT performances? We hypothesized that a learning mechanism replicating the human brain is effective for tasks where machine intelligence is difficult. Multiple experiments corresponding to specific types of MCTs where machine intelligence has room to improve performance were performed using predictive coding, a more biologically plausible learning algorithm than backpropagation. This study regarded incremental learning, long-tailed, and few-shot recognition as representative MCTs. With extensive experiments, we examined the effectiveness of predictive coding that robustly outperformed backpropagation-trained networks for the MCTs. We demonstrated that predictive coding-based incremental learning alleviates the effect of catastrophic forgetting. Next, predictive coding-based learning mitigates the classification bias in long-tailed recognition. Finally, we verified that the network trained with predictive coding could correctly predict corresponding targets with few samples. We analyzed the experimental result by drawing analogies between the properties of predictive coding networks and those of the human brain and discussing the potential of predictive coding networks in general machine learning.